Artificial Intelligence and Social Media Utilization for Rural Patients with Acute Brain Conditions in Chuncheon, Gangwon-do, South Korea

Artificial Intelligence and Social Media Utilization for Rural Patients with Acute Brain Conditions in Chuncheon, Gangwon-do, South Korea | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article Artificial Intelligence and Social Media Utilization for Rural Patients with Acute Brain Conditions in Chuncheon, Gangwon-do, South Korea Mu Seung Park, Seung-Ho Shin, Seunghun Han, Jaewoong Kang, Sang-Hwa Lee, and 19 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-6542673/v1 This work is licensed under a CC BY 4.0 License Status: Posted Version 1 posted You are reading this latest preprint version Abstract Background Despite nationwide efforts to enhance the quality of treatment for acute brain conditions in Korea, regional disparities persist due to the lack of neurology specialists and infrastructure shortcomings in rural areas. Methods We implemented two digital technologies, namely, artificial intelligence (AI)-based telemedicine and social media-based patient transfer platforms, from January 2024 to improve treatment quality for early-stage patients with various brain conditions in rural hospitals and facilitate links with regional hub hospitals. Here, we review medical records, share our experience of using digital technologies, and address current limitations and future perspectives. Results The AI-based platform was installed to facilitate collaboration between non-experts at rural hospitals and experts at hub hospitals, and the social media-based platform was adopted to improve collaboration between experts. Eight patients with a mean age of 70.7 years used the AI-based platform to facilitate accurate diagnosis and treatment. The non-experts who referred patients included general practitioners (n = 5, 62.5%), an internist (n = 1, 12.5%), and nurses (n = 2, 25.0%). The platform enabled rapid diagnosis and decision-making, and its use led to favourable outcomes. The social media-based platform was used to transfer 12 diagnosed patients. Eleven patients (91.7%) received neurocritical care, and three (25.0%) underwent surgical procedures at a hub hospital after transfer. Nine patients (75.0%) had favourable outcomes. Conclusion We suggest a novel means of reducing regional inequities in the treatment of acute brain conditions that addresses the diversity of rural medical environments. The two digital technologies implemented have helped rural hospitals respond early and facilitated inter-hospital transfer. Additional features that consider user convenience and automatic linkage of diagnosis and treatment are essential to enable the nationwide expansion of the above platforms. Telemedicine Artificial intelligence Social media Health inequity Digital technology Figures Figure 1 Figure 2 Figure 3 Figure 4 Background Acute brain conditions refer to sudden brain diseases that occur spontaneously or traumatically. Diseases that fall into this category include spontaneous intracerebral haemorrhage (ICH) with cerebral infarction (also referred to as acute stroke) and traumatic brain injury (TBI) [ 1 , 2 ]. Supporting policies and investments in medical infrastructures have been made to treat acute brain diseases in Korea over the past several decades. A nationwide population-based study showed that modern reperfusion therapy reduced the 3-month mortality rate from 4.78–3.71% [ 3 ]. Nevertheless, socioeconomic status and regional disparities influence the treatment outcomes of acute brain conditions in Korea[ 4 , 5 ]. Park et al. [ 4 ] reported that the difference in hospitalization risks due to intracranial injury was significantly greater in the most deprived than in the least deprived areas, regardless of gender or age. Also, defect-free stroke care depended on hospital size; specifically, outcomes were more favourable in hospitals with ≥ 25 stroke cases per month (odds ratio: 2.83; 95% confidence interval: 1.69–4.72) [ 5 ]. These reported differences in treatment outcomes are fundamentally attributable to regional inequalities in critical healthcare infrastructure, particularly regarding neurological care in emergency rooms (ERs). Gangwon-do is the second largest province in South Korea, but rugged mountainous terrain accounts for approximately 80% of its land area. According to Statistics Korea, Gangwon-do has the lowest population density in the country, with only 91 inhabitants per square kilometer [ 6 ]. Gangwon-do is divided into Yeongseo and Yeongdong regions by a mountain range and medically underserved rural areas [ 7 ]. The capital city of Gangwon-do, Chuncheon, is responsible for individuals with acute brain conditions residing in the neighbouring rural areas of Hongcheon, Yanggu, Inje, and Hwacheon. Gangwon-do has a stroke-related mortality rate of 33.6 per 100,000 inhabitants, which is greater than those of Seoul, Gyeonggi-do, and the national average (Fig. 1 A) [ 8 ]. These statistics suggest that accessibility and the quality of medical services determine the outcomes of emergent brain conditions. In these rural areas, there is no hospital that can perform emergent surgery for brain haemorrhage and intra-arterial (IA) thrombectomy for acute ischemic stroke with cerebral artery occlusion. Accordingly, neurological emergencies in rural ERs require that patients be transferred to Chuncheon City for emergent procedures and neurocritical care. Furthermore, from an emergency medicine perspective, it is important to keep to the golden time, that is, appropriate treatment commencement 60 minutes after symptom onset [ 9 , 10 ]. In patients with suspected acute brain conditions, rapid diagnosis and treatment while deciding on hub hospital transfer are essential to minimize neurological complications. However, the reality is that in the ERs of rural hospitals there is a severe shortage of experts capable of properly interpreting brain CT scans and providing treatment in the emergent period. Therefore, it appears these limitations underlie the higher mortality rates of acute brain conditions in the Gangwon-do region. As medical professionals working at hub hospitals responsible for nearby rural areas, we sought to address this issue by implementing two digital health solutions for emergent brain conditions: an AI-based telemedicine platform and a KakaoTalk-based web service to facilitate patient transfer. These two projects are being funded separately by government grants, and the technical components, participating medical professionals, and targeted diseases differ somewhat. Here, we describe our experience of these digital medical technologies, which were implemented to enable neighbouring rural hospitals to better treat acute brain conditions. Also, we discuss current limitations and future perspectives based on our conviction that digital applications will be instrumental in narrowing regional health disparities in areas where healthcare infrastructure for emergent brain conditions is lacking. Methods Study design This retrospective case study involved patients with an acute brain condition who presented initially at rural hospitals near Chuncheon City and were subsequently transported to a hub hospital using an AI-based telemedicine or social media platform between January 2024 and March 2025. Baseline demographic features, diagnoses, modes of onset, Glasgow Coma Scale (GCS) scores, and treatment procedures were reviewed. Favourable outcomes were defined as modified Rankin Scale (mRS) scores of 0 to 2 at 3 months after onset. The study was approved by the Institutional Research Ethics Committee of Chuncheon Sacred Heart Hospital (No. 2023-10-008). The need for informed consent was waived because of the retrospective nature of the study and the confidential record review [ 11 ]. AI-based telemedicine platform for intracranial haemorrhage Unlike Seoul or metropolitan cities, general practitioners (GPs) or non-specialists in ICH oversee ERs in most rural areas in Gangwon province. Doctors lacking ICH experience are inevitably unable to perform swift diagnoses and provide timely treatment during the acute period, which potentially leads to unfavourable neurological outcomes. To increase the treatment quality of ICH patients who visit local hospitals in rural and medically underserved areas, we initiated the development of a cloud-based platform in collaboration with the government and information technology companies in 2022 (Project Number: 2710000241, RS-2022-00155659) [ 12 , 13 ]. This platform was developed to strengthen the clinical capabilities of doctors with less ICH expertise by providing AI assistance with CT interpretation and teleconsultation (Fig. 1 B) [ 12 , 14 ]. To be more specific, a doctor at a local hospital in a rural area uploads whole CT images to the cloud, and then AI automatically determines whether a haemorrhage is present. Based on this interpretation, the doctor can request teleconsultation regarding the acute management of ICH and interhospital transfer. In addition, an ICH expert at the hub hospital can provide initial guidance on blood pressure control and securing an airway, and if a webcam is installed, neurological experts can deliver real-time remote consultation via video. Thus, this system facilitates the rapid transfer of patients requiring surgical procedures. In summary, the AI and telemedicine platform provides timely ICH diagnosis and enables proper initial treatment for spontaneous ICH and TBI and rapid transfer to a regional hub hospital for critical care and surgical treatment. In addition, the telemedicine platform can help doctors and nurses without ICH expertise in real settings. Social media-based patient transfer in acute stroke The objective of this project was to establish a network between regional hub hospitals and nearly local hospitals that cannot treat acute stroke. KakaoTalk, the most used messenger service in Korea (launched by Kakao Co. Ltd. in 2010), was used to construct the platform [ 15 ]. Briefly, the platform is operated as follows: 1) A doctor referring a stroke patient enters the KakaoTalk channel and selects from the menu a transfer request; 2) Basic patient information, vital signs, and key CT or MR images are entered; 3) A notification message is then sent to medical professionals registered on the channel; 4) Final transfer destination is determined based on considerations of recipient hospital intensive care unit and surgical availabilities. The main difference between the devised AI-based telemedicine platform and a previously reported AI-based telemedicine platform is that it is designed for patients with a diagnosis of stroke. Therefore, participants are neurosurgeons, neurologists, or emergency medicine doctors, not general doctors or nurses. The platform focuses on facilitating the transfer of stroke patients eligible for endovascular treatments or surgical procedures. The project is being conducted mainly in the cities of Chuncheon and Wonju. Results AI-based telemedicine platform Eight patients used the AI-based platform for diagnosis and treatment. Mean patient age was 70.5 years (range: 57–87), and five were men (Table 1 ). Most were diagnosed with TBI, for example, subdural or epidural haematoma. The remainder were suspected of having ICH initially, but there was no evidence of haemorrhage requiring further treatment. Senen patients were managed conservatively, and one underwent surgery. Four of the five patients transferred to a hub hospital received intensive care unit treatment. Their mean ICU stay was 2.3 days, and their hospital stays ranged from 4 and 14 days. All patients had favourable outcomes at 3 months. GPs referred six of the patients, and a nurse and an internist referred one patient apiece. Table 1 Baseline characteristics and clinical features of AI-based telemedicine and the social media-based platform for patients with acute brain conditions who present at rural hospitals Type No Age/sex Diagnosis Referral medical staff in rural hospitals Medical staff in a hub hospital GCS Management ICU stay 3-month mRS AI-based telemedicine platform 1 73/M Seizure Nurse ㅡ 15 Conservative ㅡ 0 2 63/M TBI General practitioners Neurosurgeon 15 Surgery 2 0 3 78/M TBI General practitioners Neurosurgeon 15 Conservative 0 0 4 57/M Normal finding General practitioners ㅡ 15 Conservative ㅡ 0 5 72/F TBI General practitioners Neurosurgeon 15 Conservative 2 0 6 87/F TBI Internist Neurosurgeon 15 Conservative 7 0 7 62/F Normal finding General practitioners ㅡ 15 Conservative ㅡ 0 8 72/M TBI Nurse Neurosurgeon 15 Conservative 3 0 Social media-based platform 1 57/M Haemorrhage stroke Neurosurgeon Neurosurgeon 15 Coil embolization 14 0 2 63/M Ischemic stroke Neurosurgeon Neurologist 15 Conservative 0 0 3 64/M TBI Neurosurgeon Neurosurgeon 15 Conservative 5 0 4 86/M Ischemic stroke Neurosurgeon Neurologist 13 Conservative 3 2 5 34/M TBI Neurosurgeon Neurosurgeon 15 Conservative 8 0 6 95/M Normal finding Neurosurgeon ㅡ 15 ㅡ ㅡ 0 7 66/M Haemorrhage stroke Neurosurgeon Neurosurgeon 15 Conservative 5 2 8 54/F Ischemic stroke Neurosurgeon Neurosurgeon 15 Conservative 3 1 9 64/M Ischemic stroke Emergency physician Neurologist 12 IA Thrombectomy 23 6 10 56/M Haemorrhage stroke Neurosurgeon Neurosurgeon 5 Surgery 32 4 11 72/M Haemorrhage stroke** Neurosurgeon Neurosurgeon 3 Conservative 0* 6 12 75/F TBI Neurosurgeon Neurosurgeon 15 Conservative 21 1 AI, artificial intelligence; GCS, Glasgow Coma Scale; IA, intraarterial; ICU, intensive care unit; mRS, modified Rankin Scale; TBI; traumatic brain injury ㅡ indicates that the patient was not hospitalized at a hub hospital. 0* indicates death in the emergency room of a hub hospital. Illustrations of representative cases A 63-year-old male visited a local ER for an aggravated, unbearable headache with dizziness. He had been involved in a motorcycle accident a month previously. In the absence of a doctor with ICH expertise in the ER, the duty doctor uploaded CT images to the cloud. AI embedded in the platform then identified subdural haematoma on the left side and marked this using different colours (Figs. 2 A and B). The patient was immediately transferred, with telemedicine consultation, to a regional hub hospital for surgery (Fig. 2 C). A CT scan taken 3 months later revealed almost total elimination of the haematoma (Fig. 2 D). The following case is an example of AI and teleconsultation use by an internist. An 86-year-old woman presented with progressive headache after a fall a week previously. AI identified and marked a haematoma at the site of the anterior falx (Figs. 2 E and F). At this time, the internist consulted a neurosurgeon at the hub hospital in real time by teleconsultation regarding the initial treatment and transfer (Fig. 2 E). The final example involves the detection of a small amount of subdural haemorrhage by AI in a male patient admitted to an orthopaedic unit for multiple traumas who complained of a persistent headache. He was taking anticoagulants for arterial fibrillation and valvular heart disease, and thus, the doctor was concerned about the possibility of brain haemorrhage. However, no specialist capable of diagnosing and treating cerebral haemorrhage was available. In this case, the AI platform ​​identified subdural haemorrhage (a white arrow, Fig. 2 G), which may not have been easily detected due to its proximity to bone. The patient was discharged after consulting a neurosurgeon. Social media-based patient transfer Twenty-four patients used the patient transfer platform during the study period (six months, October 2024 – March 2025), and the 12 patients admitted to a single institution in Chuncheon City were included. Mean patient age was 67.8 years, with ten men and two women. The diagnoses were ischemic stroke in 4 patients, haemorrhagic stroke in 4 patients, and TBI in 3 patients. One patient was referred for suspected cerebral infarction but discharged after a negative diffusion-weighted MRI finding. Eight patients were managed conservatively, and three patients underwent surgical procedures, including coil embolization, IA thrombectomy, or haematoma stereotactic catheter placement for haematoma aspiration (Table 1 ). Mean ICU length of stay was 10.36 days. Of the eleven survivors, seven achieved favourable functional outcomes. Illustrations of representative cases The first representative case involved a patient diagnosed with haemorrhagic stroke in the ER of a university hospital who was transferred to another university hospital for emergent surgery. A 56-year-old male with hypertension and gout presented with a stuporous mentality. Initial CT findings showed multiple haemorrhagic lesions of thalamus, ventricles, and parietal lobe (black arrows, Fig. 3 A). However, the on-duty neurosurgeon was performing surgery, and thus, an emergent procedure could not be performed immediately. The neurosurgeon inquired through the KakaoTalk platform whether the patient could be operated on immediately at another hospital, and within 5 minutes, a hospital was identified that could perform the surgery immediately. Two catheters targeting the ventricle and haematoma were placed to increase ICP control by cerebrospinal fluid and haematoma drainage. A CT scan taken three days later revealed the haematoma had decreased, which concurred with relieved ICP (Figs. 3 B and C). Two-month follow-up CT showed the haematoma had disappeared without ventricular enlargement (Fig. 3 D). At this time, the patient exhibited near-normal levels of consciousness and was transferred to a rehabilitation hospital. The second case involves the timely transfer of a patient with acute cerebral infarction to a hub hospital providing neurocritical care. An 86-year-old man visited a local ER for right-side motor weakness with diplopia of two days duration. Diffusion MRI revealed acute cerebral infarctions in the vascular territory of the posterior cerebral artery (white arrows, Fig. 3 E). Within 5 minutes of placing a transfer request on the platform, the patient was matched to a hospital capable of neurocritical care. One month later, the decline in motor performance of right extremities had improved to grade IV + from IV at the time of admission, and the infarction had not increased in extent (Fig. 3 F). The third case involved the transfer of a hemodynamically unstable patient. A 72-year-old male was found unconscious in a parking lot. He was immediately transferred to the ER of a nearby rural hospital and was diagnosed with severe brain swelling with multiple cortical SAH, SDH, and fractures of the right temporal and occipital bones with associated lesions (white arrows, Figs. 3 G and H). CPR was performed due to sudden cardiac arrest, and a request for patient transfer was made via KakaoTalk for professional intensive care management. The patient was transferred to a hub hospital without delay. Nevertheless, the patient died despite continuous CPR after transport. This case demonstrates that hemodynamically unstable patients should receive continuous monitoring and treatment by specialists during transport. Discussion Korea has made remarkable progress in the fields of economics, technology, and medicine over the last three decades. However, this rapid growth inevitably has led to differences in the distribution of regional wealth and resulted in qualitative and quantitative differences in healthcare systems [ 16 ]. Although regional differences in overall avoidable and preventable deaths in metropolitan and non-metropolitan areas were alleviated somewhat between 1995 and 2019 [ 17 ], the treatment of acute brain conditions is still subject to regional disparities [ 16 ]. Furthermore, under the conditions imposed by the COVID-19 pandemic, it became difficult to treat critically neurologically impaired patients. A nationwide web- and mobile phone-based teleconsultation network for rural and urban hospitals has been established in South Korea specifically for the treatment of emergent patients [ 18 ]. The remote emergency consultation afforded by this system achieved reductions in the unnecessary transportation of trauma patients without severe injuries [ 18 ]. Lee et al. [ 19 ] also reported that video meetings with caregivers using a mobile device increase caregiver satisfaction under the restrictions imposed due to COVID-19. However, the nationwide emergency teleconsultation network is more focused on remote imaging interpretation of overall general emergencies or the transfer of trauma patients than on patients with acute brain conditions such as stroke [ 18 ]. Furthermore, mobile phone-based video interviews for neurosurgical critically ill patients are not designed to strengthen the clinical capacity of rural hospitals to treat the acute phase of acute brain conditions [ 19 ]. The “golden hour” is an important treatment maxim for acute brain conditions, which require two completely different surgical procedures, viz., craniotomy through skull opening or endovascular treatment via femoral and radial arteries. Thus, it is important to nominate hub hospitals that can provide appropriate treatment, especially in rural areas where treatment is not possible. In addition, accurate diagnoses by medical imaging and the provision of early, timely treatment by doctors who encounter neurologically ill patients for the first time in rural ERs are often overlooked priorities. CT and MRI (magnetic resonance imaging) are primarily used to identify brain abnormalities. However, though most ERs in rural Korea have CT units, relatively few have MRI instruments. Consequently, when developing an AI-based automatic interpreting application, CT provides the best means of enhancing the diagnostic capabilities of rural ERs. Early medical treatment, like surgical or endovascular intervention, is also essential for achieving favourable neurological outcomes. In particular, the amount of haemorrhage haematoma increases within the first 6–24 hours after ictus [ 13 , 20 , 21 ]. Arima et al [ 22 ]. reported that intensive blood pressure (BP) lowering achieved benefits in patients with a neurological condition, notably, to between a systolic BP of 130–139 mm Hg in patients with acute ICH. In addition to BP control, appropriate airway, antiepileptic drug use, and increased intracranial pressure (IICP) management should be performed simultaneously when ICH is diagnosed, as these favourably influence neurological outcomes [ 13 , 23 ]. However, the reality is that most doctors working in rural ERs in Korea are GPs, and even among specialist doctors, few are experts in acute brain conditions. Thus, swift diagnosis and timely appropriate treatment cannot be conducted effectively. Digital health technologies are increasingly being used to transform care paradigms for patients with acute brain conditions [ 24 ]. In particular, the appropriate application of AI, mobile and wireless technologies, and cloud platform services could potentially improve treatment capabilities in rural hospitals through telemedicine. In the United States, so-called telestroke services are being actively utilized to diagnose ischemic stroke patients indicated for thrombolysis or endovascular thrombectomy rapidly. Zachrison et al [ 25 ]. reported that consultation start time and thrombolysis performance time decreased over time after introducing a telestroke platform. Also, telestroke significantly increased the rate of receiving treatment within 3 hours of symptom onset (OR = 2.15; 95% CI 1.37–3.40), improved neurological outcomes at 3 months (OR = 1.29; 95% CI 1.01–1.63), and lowered the in-hospital mortality rate (OR = 0.67; 95% CI 0.52–0.87) [ 26 ]. Ford et al. [ 27 ] revealed that telemedicine assessment reduced the times required to achieve BP control and anticoagulation reversal. Accordingly, the application of digital technologies in rural areas is likely associated with better prognoses in patients with an acute brain condition [ 26 ]. However, telemedicine is still not generally utilized due to regulations regarding patients with acute brain conditions, although the Korean government, prompted by the COVID-19 pandemic, temporarily allowed telemedicine [ 19 ]. Fortunately, Gangwon-do has been designated a regulatory-free zone by the Korean government, which has made telemedicine possible. Here, we implemented two platform types to address the special characteristics of Gangwon-do province. To design these platforms, we focused on the expertise of initial care physicians in rural hospitals and their requirements from neurological specialists in hub hospitals. For example, if a doctor does not have specialized knowledge about neurological abnormalities, AI could be used to aid the diagnosis and provide details of initial medical requirements such as BP control, respiration stabilization, and IICP management before transfer. Inexperienced doctors often have difficulty differentiating normal and abnormal conditions, and in such cases, AI interpretation with subsequent teleconsultation can aid diagnosis prior to transfer. Concern remains that the rapid pace of AI advancement may outpace the infrastructure capabilities of rural regions and pose a barrier to sustainable implementation. This issue can be addressed through a cloud-based platform, which offers a flexible environment for the implementation of artificial intelligence. On the other hand, when referring physicians at rural hospitals are neurosurgeons, neurologists, or emergency medicine specialists capable of diagnosing brain diseases by CT or MRI and initiating early medical treatment, the main issue is to transfer the patient to a hub as quickly as possible for surgical treatment if needed. Social media-based platforms enable effective communication between experts, and a social network approach enhances accessibility by leveraging a mobile communication platform used by over 95% of smartphone users in Korea. Nevertheless, further research is needed to determine whether large amounts of medical data can be processed reliably and stably in emergent situations. While operating the artificial intelligence (AI)-based telemedicine and social media-based patient transfer platforms, we identified two issues that need to be addressed to maximize their effectiveness in medical settings. The first issue is to improve the effectiveness of cooperation, particularly for unstable patients with acute, severe brain conditions. The majority of cases subjected to digital technologies and transferred to a hub hospital did not have severe or unstable neurological complications. However, even when digital technology was applied, patients with acute and severe brain disease were transferred to a hub hospital from areas served by rural hospitals. We suppose such transfers are due to a fear of being presented with seriously ill patients. This fear can only be resolved by education, and we suggest virtual reality (VR) may be suitable in emergent and unstable situations. From the perspective of the learner, VR conveys complex issues quickly and has the potential to replace mannequin-based courses [ 28 ]. Accordingly, additional educational programs on severe neurological disorders should be provided to staff in rural ERs to ensure proper neurological examinations, diagnosis, and transportation. The second issue is integration with an AI platform capable of diagnosing ischemic stroke. According to Korean statistics, ischemic stroke accounts for about 70% of all stroke cases [ 29 ]. To facilitate the use of tissue-type plasminogen activator (tPA) in rural ERs, a CT-based AI algorithm to identify ischemic stroke is needed. Moreover, for effective management in rural ERs, it is necessary not only to have a highly accurate diagnostic algorithm for ischemic stroke, but also to implement functions that enable integration with neuromonitoring and prognosis estimation systems [ 30 ]. In this study, we applied two different digital technologies in real-world clinical settings to improve rural healthcare environments and evaluated their impacts on the quality of treatment afforded to acute brain conditions. Through the indirect integration of the two platforms, we gained insights into how a more effective telemedicine system could be structured (Fig. 4 ). Integrating different platforms allows individual telemedicine systems to be utilized beyond their intended purposes. However, such integration requires careful consideration to mitigate unforeseen risks associated with unintended use. The application of the most recent medical AI systems provides a means of addressing this challenge. The integration of medical AI and telemedicine can provide AI-assistant tools that help correct diagnostic errors by non-experts and generative AI systems. Recent studies have reported that AI-powered diagnostic support tools demonstrated excellent performance when assisting non-task-specific expert clinicians or non-experienced healthcare providers [ 31 , 32 ]. At present, generative AI requires the establishment of safety regulations and standardization to ensure reliability, but it has substantial potential for broader applications [ 33 ]. In particular, large language models (LLMs) offer better interpretations of complex medical terminology and summaries of large volumes of clinical information [ 34 , 35 ]. In addition to providing early diagnosis, LLMs offer comprehensive and detailed guidance on treatment options and use hospital-specific terminology or contextual factors in the guidance process. By leveraging relatively flexible data structures, generative AI enables the expansion of telemedicine platforms into home-based rehabilitation through technologies such as the Internet of Things (IoT) and is anticipated to play a pivotal role in the future development of telemedicine. Conclusions AI-based telemedicine and social media-based transfer platforms offer a promising solution to the practical challenges posed by limited medical infrastructure in rural areas. Their implementation facilitates early diagnosis and appropriate treatment for patients with acute brain conditions and provides efficient interhospital transfers regardless of the level of the initial hospital. These findings suggest that long-term, innovative, digital health projects are essential for securing treatment enhancements in the future rural medical environment. Declarations Ethics approval and consent to participate This study was approved by the institutional review board of Hallym University Chuncheon Sacred Heart Hospital (Approval number. IRB 2021-10-012-007). All experimental methods complied with the Helsinki Declaration. The need for informed consent was waived by the IRB of our hospital due to the retrospective design of the study. Consent for publication Not applicable. Data availability Please direct enquiries to the corresponding author. Conflicts of interest The authors have no conflict of interest to declare. Funding This work was supported by a Korea Medical Device Development Fund grant funded by the Ministry of Science and ICT, the Ministry of Trade, Industry and Energy, the Ministry of Health & Welfare, the Ministry of Food and Drug Safety (Project Number: 2710000241, RS-2022-00155659), the Pilot Project for Severe and Emergency Cardiovascular Disease Problem-Solving Treatment Cooperation Network, the Hallym University Medical Center Research Fund, and the Hallym University Research Fund (HURF). Authors' Contributions MSP and S-HS contributed to data curation, formal analysis, and drafting of the original manuscript. JPJ and CK were involved in conceptualization, methodology development, formal analysis, investigation, funding acquisition, and manuscript drafting. SH, JK, S-HL, J-KR, SJ, SHS, JEK, J-HS, HJC, HSJ, JHA, SJL, SK, JJL, YHI, HJ, HK, JSY, SHK, and YJC contributed to data collection, data validation, and resource provision. All authors reviewed and approved the final version of the manuscript for submission. References Jeon JP, Lee SU, Kim SE, Kang SH, Yang JS, Choi HJ et al. Correlation of optic nerve sheath diameter with directly measured intracranial pressure in Korean adults using bedside ultrasonography. PLoS One. 2017;12(9):e0183170. PMID: 28902893. 10.1371/journal.pone.0183170 Vespa P. Continuous EEG monitoring for the detection of seizures in traumatic brain injury, infarction, and intracerebral hemorrhage: to detect and protect. J Clin Neurophysiol. 2005;22(2):99–106. 10.1097/01.wnp.0000154919.54202.e0 . PMID: 15805809. Park YK, Yoon BH, Won YD, Kim JH, Kang HI. Real-World Impact of Modern Reperfusion Therapy for Acute Ischemic Stroke: A Nationwide Population-Based Data Study in Korea. J Korean Neurosurg Soc. 2024;67(2):186–93. 10.3340/jkns.2023.0133 . PMID: 37799025. Park HA, Vaca FE, Jung-Choi K, Park H, Park JO. Area-Level Socioeconomic Inequalities in Intracranial Injury-Related Hospitalization. J Korean Med Sci. 2023;38(4):e38. 10.3346/jkms.2023.38.e38 . PMID: 36718564. in Korea: A Retrospective Analysis of Data From Korea National Hospital Discharge Survey 2008–2015. Park HK, Kim SE, Cho YJ, Kim JY, Oh H, Kim BJ, et al. Quality of acute stroke care in Korea (2008–2014): Retrospective analysis of the nationwide and nonselective data for quality of acute stroke care. Eur Stroke J. 2019;4(4):337–46. 10.1177/2396987319849983 . PMID: 31903432. Korea S, e-Nara I. 2025 March 24; https://www.index.go.kr/unity/potal/main/EachDtlPageDetail.do?idx_cd=1007 Kim YJ, Li L, Hwang JY. A Maternity Waiting Home Is an Alternative Approach for the Accessibility of Pregnant Women in an Obstetrically Underserved Area of Korea. J Korean Med Sci. 2023;38(17):e164. 10.3346/jkms.2023.38.e164 . PMID: 37128881. Kim JY, Kang K, Kang J, Koo J, Kim DH, Kim BJ, et al. Executive Summary of Stroke Statistics in Korea 2018: A Report from the Epidemiology Research Council of the Korean Stroke Society. J Stroke. 2019;21(1):42–59. 10.5853/jos.2018.03125 . PMID: 30558400. Kim JT, Fonarow GC, Smith EE, Reeves MJ, Navalkele DD, Grotta JC, et al. Treatment With Tissue Plasminogen Activator in the Golden Hour and the Shape of the 4.5-Hour Time-Benefit Curve in the National United States Get With The Guidelines-Stroke Population. Circulation. 2017;135(2):128–39. 10.1161/CIRCULATIONAHA.116.023336 . PMID: 27815374. Randhawa AS, Pariona-Vargas F, Starkman S, Sanossian N, Liebeskind DS, Avila G, et al. Beyond the Golden Hour: Treating Acute Stroke in the Platinum 30 Minutes. Stroke. 2022;53(8):2426–34. 10.1161/STROKEAHA.121.036993 . PMID: 35545939. Misirlioglu M, Ekinci F, Yildizdas D, Horoz OO, Yilmaz HL, Incecik F, et al. A Retrospective Cohort Study of Traumatic Brain Injury in Children: A Single-Institution Experience and Determinants of Neurologic Outcome. J Crit Care Med (Targu Mures). 2023;9(4):252–61. 10.2478/jccm-2023-0027 . PMID: 37969881. Jun HS, Yang K, Kim J, Jeon JP, Ahn JH, Lee SJ, et al. Development of Cloud-Based Telemedicine Platform for Acute Intracerebral Hemorrhage in Gangwon-do: Concept and Protocol. J Korean Neurosurg Soc. 2023;66(5):488–93. 10.3340/jkns.2022.0256 . PMID: 36756670. Jun HS, Yang K, Kim J, Jeon JP, Kim SJ, Ahn JH, et al. Telemedicine Protocols for the Management of Patients with Acute Spontaneous Intracerebral Hemorrhage in Rural and Medically Underserved Areas in Gangwon State: Recommendations for Doctors with Less Expertise at Local Emergency Rooms. J Korean Neurosurg Soc. 2024;67(4):385–96. 10.3340/jkns.2023.0199 . PMID: 37901932. Yun TJ, Choi JW, Han M, Jung WS, Choi SH, Yoo RE et al. Deep learning based automatic detection algorithm for acute intracranial haemorrhage: a pivotal randomized clinical trial. NPJ Digit Med. 2023;6(1):61. PMID: 37029272. 10.1038/s41746-023-00798-8 Woo SH, Oh EG, Kim KS, Chu SH, Kim GS, Nam CM. Development and Assessment of a Social Network Service-Based Lifestyle-Modification Program for Workers at High Risk of Developing Cardiovascular Disease. Workplace Health Saf. 2020;68(3):109–20. 10.1177/2165079919864976 . PMID: 31434552. Jung EJ, Kim DY, Bae HJ, Ko KP. Assessing regional disparities and vulnerability in stroke care across Gyeonggi Province: A focus on hospital service areas. J Stroke Cerebrovasc Dis. 2024;33(9):107817. 10.1016/j.jstrokecerebrovasdis.2024.107817 . PMID: 38880365. Choi MH, Moon MH, Yoon TH. Avoidable Mortality between Metropolitan and Non-Metropolitan Areas in Korea from 1995 to 2019: A Descriptive Study of Implications for the National Healthcare Policy. Int J Environ Res Public Health. 2022;19(6). PMID: 35329162. 10.3390/ijerph19063475 Choi W, Lim Y, Heo T, Lee S, Kim W, Kim SC, et al. Characteristics and Effectiveness of Mobile- and Web-Based Tele-Emergency Consultation System between Rural and Urban Hospitals in South Korea: A National-Wide Observation Study. J Clin Med. 2023;12(19). 10.3390/jcm12196252 . PMID: 37834896. Lee MH, Jang SR, Lee TK. The Direction of Neurosurgery to Overcome the Living with COVID-19 Era: The Possibility of Telemedicine in Neurosurgery. J Korean Neurosurg Soc. 2023;66(5):573–81. 10.3340/jkns.2022.0211 . PMID: 37667635. Brouwers HB, Greenberg SM. Hematoma expansion following acute intracerebral hemorrhage. Cerebrovasc Dis. 2013;35(3):195–201. 10.1159/000346599 . PMID: 23466430. Kuohn LR, Witsch J, Steiner T, Sheth KN, Kamel H, Navi BB, et al. Early Deterioration, Hematoma Expansion, and Outcomes in Deep Versus Lobar Intracerebral Hemorrhage: The FAST Trial. Stroke. 2022;53(8):2441–8. 10.1161/STROKEAHA.121.037974 . PMID: 35360929. Arima H, Heeley E, Delcourt C, Hirakawa Y, Wang X, Woodward M, et al. Optimal achieved blood pressure in acute intracerebral hemorrhage: INTERACT2. Neurology. 2015;84(5):464–71. 10.1212/WNL.0000000000001205 . PMID: 25552575. Li Z, You M, Long C, Bi R, Xu H, He Q et al. Hematoma Expansion in Intracerebral Hemorrhage: An Update on Prediction and Treatment. Front Neurol. 2020;11:702. PMID: 32765408. 10.3389/fneur.2020.00702 Silva GS, Andrade JBC. Digital health in stroke: a narrative review. Arq Neuropsiquiatr. 2024;82(8):1–10. 10.1055/s-0044-1789201 . PMID: 39187259. Zachrison KS, Sharma R, Wang Y, Mehrotra A, Schwamm LH. National Trends in Telestroke Utilization in a US Commercial Platform Prior to the COVID-19 Pandemic. J Stroke Cerebrovasc Dis. 2021;30(10):106035. 10.1016/j.jstrokecerebrovasdis.2021.106035 . PMID: 34419836. Lazarus G, Permana AP, Nugroho SW, Audrey J, Wijaya DN, Widyahening IS. Telestroke strategies to enhance acute stroke management in rural settings: A systematic review and meta-analysis. Brain Behav. 2020;10(10):e01787. PMID: 32812380. 10.1002/brb3.1787 Ford S, Ajani Z, Chen Q, Sorreda V, Tu G, McCartney D, et al. Comparison of Standard Emergency Room Care with Tele-Stroke Evaluation in Acute Intracerebral Hemorrhage Management (P6. 030). Neurology. 2016;86(16supplement):P6. Mahling M, Wunderlich R, Steiner D, Gorgati E, Festl-Wietek T, Herrmann-Werner A. Virtual Reality for Emergency Medicine Training in Medical School: Prospective, Large-Cohort Implementation Study. J Med Internet Res. 2023;25:e43649. 10.2196/43649 . PMID: 36867440. Hong KS, Bang OY, Kim JS, Heo JH, Yu KH, Bae HJ, et al. Stroke Statistics in Korea: Part II Stroke Awareness and Acute Stroke Care, A Report from the Korean Stroke Society and Clinical Research Center For Stroke. J Stroke. 2013;15(2):67–77. 10.5853/jos.2013.15.2.67 . PMID: 24324942. Uparela-Reyes MJ, Villegas-Trujillo LM, Cespedes J, Velasquez-Vera M, Rubiano AM. Usefulness of Artificial Intelligence in Traumatic Brain Injury: A Bibliometric Analysis and Mini-review. World Neurosurg. 2024;188:83–92. 10.1016/j.wneu.2024.05.065 . PMID: 38759786. Yanagawa M. Artificial Intelligence Improves Radiologist Performance for Predicting Malignancy at Chest CT. Radiology. 2022;304(3):692–3. 10.1148/radiol.220571 . PMID: 35608448. Gaube S, Suresh H, Raue M, Lermer E, Koch TK, Hudecek MFC et al. Non-task expert physicians benefit from correct explainable AI advice when reviewing X-rays. Sci Rep. 2023;13(1):1383. PMID: 36697450. 10.1038/s41598-023-28633-w Goodman KE, Yi PH, Morgan DJ. AI-Generated Clinical Summaries Require More Than Accuracy. JAMA. 2024;331(8):637–8. 10.1001/jama.2024.0555 . PMID: 38285439. Shool S, Adimi S, Saboori Amleshi R, Bitaraf E, Golpira R, Tara M. A systematic review of large language model (LLM) evaluations in clinical medicine. BMC Med Inform Decis Mak. 2025;25(1):117. PMID: 40055694. 10.1186/s12911-025-02954-4 Wals Zurita AJ, Miras Del Rio H, Ugarte Ruiz de Aguirre N, Nebrera Navarro C, Rubio Jimenez M, Munoz Carmona D, et al. The Transformative Potential of Large Language Models in Mining Electronic Health Records Data: Content Analysis. JMIR Med Inf. 2025;13:e58457. 10.2196/58457 . PMID: 39746191. Additional Declarations No competing interests reported. Cite Share Download PDF Status: Posted Version 1 posted You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-6542673","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":457626811,"identity":"755cc8d5-6177-4fdd-bfb3-0337cbd67a5a","order_by":0,"name":"Mu Seung Park","email":"","orcid":"","institution":"Hallym University College of Medicine","correspondingAuthor":false,"prefix":"","firstName":"Mu","middleName":"Seung","lastName":"Park","suffix":""},{"id":457626812,"identity":"7f4217c1-72d9-4fb2-b13a-98249f1a6752","order_by":1,"name":"Seung-Ho Shin","email":"","orcid":"","institution":"Hallym University Sacred Heart Hospital","correspondingAuthor":false,"prefix":"","firstName":"Seung-Ho","middleName":"","lastName":"Shin","suffix":""},{"id":457626813,"identity":"fd1cdb68-e8b7-4be2-9ff9-d6446066a38e","order_by":2,"name":"Seunghun Han","email":"","orcid":"","institution":"Hallym University College of Medicine","correspondingAuthor":false,"prefix":"","firstName":"Seunghun","middleName":"","lastName":"Han","suffix":""},{"id":457626814,"identity":"bced5158-8683-4f14-9798-f2232a960be2","order_by":3,"name":"Jaewoong Kang","email":"","orcid":"","institution":"Hallym University Sacred Heart Hospital","correspondingAuthor":false,"prefix":"","firstName":"Jaewoong","middleName":"","lastName":"Kang","suffix":""},{"id":457626815,"identity":"e3c978e7-3c68-4c97-94b8-2a672aabf6cf","order_by":4,"name":"Sang-Hwa Lee","email":"","orcid":"","institution":"Hallym University College of Medicine","correspondingAuthor":false,"prefix":"","firstName":"Sang-Hwa","middleName":"","lastName":"Lee","suffix":""},{"id":457626816,"identity":"1137bcab-ff9c-4b42-a435-9e72be0fbd7d","order_by":5,"name":"Jong-Kook Rhim","email":"","orcid":"","institution":"Jeju National University College of Medicine","correspondingAuthor":false,"prefix":"","firstName":"Jong-Kook","middleName":"","lastName":"Rhim","suffix":""},{"id":457626817,"identity":"ecedbb06-664a-41b3-8d1b-07c9f1446a4e","order_by":6,"name":"Sungpil Joo","email":"","orcid":"","institution":"Chonnam National University Hospital \u0026 Medical School","correspondingAuthor":false,"prefix":"","firstName":"Sungpil","middleName":"","lastName":"Joo","suffix":""},{"id":457626818,"identity":"16ef42cc-4912-45eb-8fef-0406b0b3e582","order_by":7,"name":"Seung Hun Sheen","email":"","orcid":"","institution":"CHA Bundang Medical Center","correspondingAuthor":false,"prefix":"","firstName":"Seung","middleName":"Hun","lastName":"Sheen","suffix":""},{"id":457626819,"identity":"b37a36f2-01ef-4522-8280-fd6d426cd616","order_by":8,"name":"Jeong Eun Kim","email":"","orcid":"","institution":"Seoul National University College of Medicine","correspondingAuthor":false,"prefix":"","firstName":"Jeong","middleName":"Eun","lastName":"Kim","suffix":""},{"id":457626820,"identity":"45789b63-b4fa-408a-9f8d-436e4b574ec4","order_by":9,"name":"Jong-Hee Sohn","email":"","orcid":"","institution":"Hallym University College of Medicine","correspondingAuthor":false,"prefix":"","firstName":"Jong-Hee","middleName":"","lastName":"Sohn","suffix":""},{"id":457626821,"identity":"972d5c5a-939e-447b-83c5-64da11ba0db1","order_by":10,"name":"Hyuk Jai Choi","email":"","orcid":"","institution":"Hallym University College of Medicine","correspondingAuthor":false,"prefix":"","firstName":"Hyuk","middleName":"Jai","lastName":"Choi","suffix":""},{"id":457626822,"identity":"9c300e58-1b6b-4833-bba4-5f58d0a88209","order_by":11,"name":"Hyo Sub Jun","email":"","orcid":"","institution":"Kangwon National University College of Medicine","correspondingAuthor":false,"prefix":"","firstName":"Hyo","middleName":"Sub","lastName":"Jun","suffix":""},{"id":457626823,"identity":"3d675022-85ec-4b13-b943-980880f6db5a","order_by":12,"name":"Jun Hyong Ahn","email":"","orcid":"","institution":"Kangwon National University College of Medicine","correspondingAuthor":false,"prefix":"","firstName":"Jun","middleName":"Hyong","lastName":"Ahn","suffix":""},{"id":457626824,"identity":"d0ec99f4-3c70-4241-a802-41faee5b1617","order_by":13,"name":"Seung Jin Lee","email":"","orcid":"","institution":"Kangwon National University College of Medicine","correspondingAuthor":false,"prefix":"","firstName":"Seung","middleName":"Jin","lastName":"Lee","suffix":""},{"id":457626825,"identity":"33e0be6f-b293-4d17-8a56-458d23920899","order_by":14,"name":"Seongheon Kim","email":"","orcid":"","institution":"Kangwon National University College of Medicine","correspondingAuthor":false,"prefix":"","firstName":"Seongheon","middleName":"","lastName":"Kim","suffix":""},{"id":457626826,"identity":"e1cd4689-b93b-4aae-8ac1-f4be01ae2dbe","order_by":15,"name":"Jae-Jun Lee","email":"","orcid":"","institution":"Hallym University College of Medicine","correspondingAuthor":false,"prefix":"","firstName":"Jae-Jun","middleName":"","lastName":"Lee","suffix":""},{"id":457626827,"identity":"cb8b20c6-7933-435f-a76d-d9c41e7cff84","order_by":16,"name":"Yong-Ho In","email":"","orcid":"","institution":"Hallym University College of Medicine","correspondingAuthor":false,"prefix":"","firstName":"Yong-Ho","middleName":"","lastName":"In","suffix":""},{"id":457626828,"identity":"3be696f8-be89-4062-9a1c-8e1a8a42134c","order_by":17,"name":"Harry Jung","email":"","orcid":"","institution":"Hallym University College of Medicine","correspondingAuthor":false,"prefix":"","firstName":"Harry","middleName":"","lastName":"Jung","suffix":""},{"id":457626829,"identity":"5d0c3cd2-a824-4326-bf68-e7f456f1673e","order_by":18,"name":"Hyesook Kim","email":"","orcid":"","institution":"Hallym University College of Medicine","correspondingAuthor":false,"prefix":"","firstName":"Hyesook","middleName":"","lastName":"Kim","suffix":""},{"id":457626830,"identity":"246856d2-b295-49a8-8c3f-88e03f97d626","order_by":19,"name":"Jin Seo Yang","email":"","orcid":"","institution":"Hallym University College of Medicine","correspondingAuthor":false,"prefix":"","firstName":"Jin","middleName":"Seo","lastName":"Yang","suffix":""},{"id":457626831,"identity":"d765562b-395d-46fb-b746-d4534df8844a","order_by":20,"name":"Suk Hyung Kang","email":"","orcid":"","institution":"Hallym University College of Medicine","correspondingAuthor":false,"prefix":"","firstName":"Suk","middleName":"Hyung","lastName":"Kang","suffix":""},{"id":457626832,"identity":"56ac933c-6512-4e01-8631-43adb685ba6f","order_by":21,"name":"Yong Jun Cho","email":"","orcid":"","institution":"Hallym University College of Medicine","correspondingAuthor":false,"prefix":"","firstName":"Yong","middleName":"Jun","lastName":"Cho","suffix":""},{"id":457626833,"identity":"5dd44116-8514-4b74-aac0-21be538a70c2","order_by":22,"name":"Chulho Kim","email":"","orcid":"","institution":"Hallym University College of Medicine","correspondingAuthor":false,"prefix":"","firstName":"Chulho","middleName":"","lastName":"Kim","suffix":""},{"id":457626835,"identity":"d2fac8fc-fcc2-4a02-881a-bd3ebd09c6b1","order_by":23,"name":"Jin Pyeong Jeon","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAAyklEQVRIiWNgGAWjYBACxoYDjI9//qmRYzhAghZmY8aGY8YwLRLE6GITZmxgTmwgWgtz4/FnzIU72NL7jp8x3cC4w6aOCIedMXs884xM7swzOWY3GM+kEbYFqIXdgIeNLXfDAZCWtsPEaDn+TIKHjTnd4PwbkJb/xGg5YCbN28acYHADbMsBohxmbDjjzDHDmTeeld1IbEuWbCCkxXDG8YcPPlTUyPOdT95242ObHT9BWwxnHEDiJRBUDwTy/AQdMgpGwSgYBSMeAAAr6EZ9MlDZzgAAAABJRU5ErkJggg==","orcid":"","institution":"Hallym University College of Medicine","correspondingAuthor":true,"prefix":"","firstName":"Jin","middleName":"Pyeong","lastName":"Jeon","suffix":""}],"badges":[],"createdAt":"2025-04-27 23:53:09","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-6542673/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-6542673/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":83284475,"identity":"6ecbc1c9-5080-43f0-9aef-398ab06c1820","added_by":"auto","created_at":"2025-05-22 11:03:08","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":370621,"visible":true,"origin":"","legend":"\u003cp\u003eRetrospective case series on the use of artificial intelligence-based telemedicine and social media-based transfer platforms for patients with acute brain conditions in rural hospitals managed by a hub hospital in Chuncheon City (Gangwon-do, South Korea). \u003cstrong\u003e(A)\u003c/strong\u003eDetailed description of locations, health care infrastructure, and stroke-related mortality rates. \u003cstrong\u003e(B and C)\u003c/strong\u003e Schematic diagrams of the two digital technologies and their treatment-strengthening abilities for acute brain conditions in rural hospitals. CT, computed tomography; ER, emergency room.\u003c/p\u003e","description":"","filename":"image1.png","url":"https://assets-eu.researchsquare.com/files/rs-6542673/v1/548dd56f09a0e217b05bcbe3.png"},{"id":83285103,"identity":"ad8696dc-0859-4d42-b330-599a6110aa03","added_by":"auto","created_at":"2025-05-22 11:11:08","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":658438,"visible":true,"origin":"","legend":"\u003cp\u003eRepresentative cases of the artificial intelligence (AI)-based telemedicine platform helping non-experts treat acute brain conditions in rural areas. \u003cstrong\u003e(A and B)\u003c/strong\u003e Coloured AI interpretation of subdural haematoma. \u003cstrong\u003e(C and D) \u003c/strong\u003eThe patient was transferred to a regional hub hospital for surgery. A CT scan taken 3 months later revealed almost elimination of the left-sided haematoma. \u003cstrong\u003e(E and F) \u003c/strong\u003eAn\u003cstrong\u003e \u003c/strong\u003einternist used AI and teleconsultation for an 86-year-old woman who presented with a progressive headache after trauma. AI helped the internist diagnose haematoma at the site of the anterior falx \u003cstrong\u003e(G)\u003c/strong\u003e, and a teleconsultation between the internist and a neurosurgeon in a hub hospital followed. This case showed AI also enabled a non-expert to detect a small amount of haemorrhage adjacent to bone.\u003c/p\u003e","description":"","filename":"image2.png","url":"https://assets-eu.researchsquare.com/files/rs-6542673/v1/2b7ccb451bd557156f8a7e8e.png"},{"id":83284482,"identity":"1f72cf96-299d-4b9c-95c9-e462658c2e8e","added_by":"auto","created_at":"2025-05-22 11:03:08","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":756530,"visible":true,"origin":"","legend":"\u003cp\u003eRepresentative cases of social media-based patient transfer enabling communications between experts. \u003cstrong\u003e(A)\u003c/strong\u003e A 56-year-old man was diagnosed with intracerebral haemorrhage (ICH), but immediate surgery was not possible. However, a neurosurgeon available for surgery was located, and the patient received appropriate treatment at another hospital after transfer. \u003cstrong\u003e(B and C) \u003c/strong\u003eA\u003cstrong\u003e \u003c/strong\u003eCT scan taken three days postoperatively showed decreased haemorrhage, which was corroborated by ameliorated intracranial pressure. \u003cstrong\u003e(D)\u003c/strong\u003e Two months later, CT revealed the haematoma had disappeared. \u003cstrong\u003e(E and F) \u003c/strong\u003eAn 86-year-old man was diagnosed with acute cerebral infarction by diffusion MRI at a rural hospital and admitted to the neurocritical care unit of a hub hospital through the platform. The patient was discharged with improved motor performance and without an infarction extent increase. \u003cstrong\u003e(G and H) \u003c/strong\u003eA 72-year-old man who visited a rural emergency department showed severe brain swelling with multiple haemorrhages and skull fractures. Cardiopulmonary resuscitation was performed, and a request for patient transfer was performed via the platform. The patient was immediately transferred to a hub hospital.\u003c/p\u003e","description":"","filename":"image3.png","url":"https://assets-eu.researchsquare.com/files/rs-6542673/v1/49ecf9de63d78a2985b9fff1.png"},{"id":83285402,"identity":"8a186715-2869-450e-afd9-878daa80b434","added_by":"auto","created_at":"2025-05-22 11:19:08","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":338452,"visible":true,"origin":"","legend":"\u003cp\u003eFuture development direction for enhancing the ability to treat acute severe brain conditions in rural areas. The platform does not require that users have specific qualifications regardless of experience or technical skills. Artificial intelligence (AI) algorithms embedded in the platform should be improved to provide optimal treatment based on clinical features and vital signs in addition to image interpretation. Multi-regional networking between rural hospitals and hub hospitals should be performed through the platform to provide nationwide expansion. Also, generative large language model (LLM)-based AI is required to ensure simultaneous treatment after diagnosis in a user-friendly, cloud-driven Internet of Things (IOT) environment.\u003c/p\u003e","description":"","filename":"image4.png","url":"https://assets-eu.researchsquare.com/files/rs-6542673/v1/3f52acd82034d77794cbc427.png"},{"id":100786558,"identity":"400ca74c-cf0c-4e8b-a64e-247ce84f9f84","added_by":"auto","created_at":"2026-01-21 11:59:45","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":3410700,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-6542673/v1/c0cf4033-2ad8-4fb3-aadf-f016aeb319a7.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"Artificial Intelligence and Social Media Utilization for Rural Patients with Acute Brain Conditions in Chuncheon, Gangwon-do, South Korea","fulltext":[{"header":"Background","content":"\u003cp\u003eAcute brain conditions refer to sudden brain diseases that occur spontaneously or traumatically. Diseases that fall into this category include spontaneous intracerebral haemorrhage (ICH) with cerebral infarction (also referred to as acute stroke) and traumatic brain injury (TBI) [\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e, \u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e]. Supporting policies and investments in medical infrastructures have been made to treat acute brain diseases in Korea over the past several decades. A nationwide population-based study showed that modern reperfusion therapy reduced the 3-month mortality rate from 4.78\u0026ndash;3.71% [\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e]. Nevertheless, socioeconomic status and regional disparities influence the treatment outcomes of acute brain conditions in Korea[\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e, \u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e]. Park et al. [\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e] reported that the difference in hospitalization risks due to intracranial injury was significantly greater in the most deprived than in the least deprived areas, regardless of gender or age. Also, defect-free stroke care depended on hospital size; specifically, outcomes were more favourable in hospitals with \u0026ge;\u0026thinsp;25 stroke cases per month (odds ratio: 2.83; 95% confidence interval: 1.69\u0026ndash;4.72) [\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e]. These reported differences in treatment outcomes are fundamentally attributable to regional inequalities in critical healthcare infrastructure, particularly regarding neurological care in emergency rooms (ERs).\u003c/p\u003e \u003cp\u003eGangwon-do is the second largest province in South Korea, but rugged mountainous terrain accounts for approximately 80% of its land area. According to Statistics Korea, Gangwon-do has the lowest population density in the country, with only 91 inhabitants per square kilometer [\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e]. Gangwon-do is divided into Yeongseo and Yeongdong regions by a mountain range and medically underserved rural areas [\u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e]. The capital city of Gangwon-do, Chuncheon, is responsible for individuals with acute brain conditions residing in the neighbouring rural areas of Hongcheon, Yanggu, Inje, and Hwacheon. Gangwon-do has a stroke-related mortality rate of 33.6 per 100,000 inhabitants, which is greater than those of Seoul, Gyeonggi-do, and the national average (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eA) [\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e]. These statistics suggest that accessibility and the quality of medical services determine the outcomes of emergent brain conditions. In these rural areas, there is no hospital that can perform emergent surgery for brain haemorrhage and intra-arterial (IA) thrombectomy for acute ischemic stroke with cerebral artery occlusion. Accordingly, neurological emergencies in rural ERs require that patients be transferred to Chuncheon City for emergent procedures and neurocritical care. Furthermore, from an emergency medicine perspective, it is important to keep to the golden time, that is, appropriate treatment commencement 60 minutes after symptom onset [\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e, \u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e]. In patients with suspected acute brain conditions, rapid diagnosis and treatment while deciding on hub hospital transfer are essential to minimize neurological complications. However, the reality is that in the ERs of rural hospitals there is a severe shortage of experts capable of properly interpreting brain CT scans and providing treatment in the emergent period. Therefore, it appears these limitations underlie the higher mortality rates of acute brain conditions in the Gangwon-do region.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eAs medical professionals working at hub hospitals responsible for nearby rural areas, we sought to address this issue by implementing two digital health solutions for emergent brain conditions: an AI-based telemedicine platform and a KakaoTalk-based web service to facilitate patient transfer. These two projects are being funded separately by government grants, and the technical components, participating medical professionals, and targeted diseases differ somewhat. Here, we describe our experience of these digital medical technologies, which were implemented to enable neighbouring rural hospitals to better treat acute brain conditions. Also, we discuss current limitations and future perspectives based on our conviction that digital applications will be instrumental in narrowing regional health disparities in areas where healthcare infrastructure for emergent brain conditions is lacking.\u003c/p\u003e"},{"header":"Methods","content":"\u003cdiv id=\"Sec3\" class=\"Section2\"\u003e \u003ch2\u003eStudy design\u003c/h2\u003e \u003cp\u003eThis retrospective case study involved patients with an acute brain condition who presented initially at rural hospitals near Chuncheon City and were subsequently transported to a hub hospital using an AI-based telemedicine or social media platform between January 2024 and March 2025. Baseline demographic features, diagnoses, modes of onset, Glasgow Coma Scale (GCS) scores, and treatment procedures were reviewed. Favourable outcomes were defined as modified Rankin Scale (mRS) scores of 0 to 2 at 3 months after onset. The study was approved by the Institutional Research Ethics Committee of Chuncheon Sacred Heart Hospital (No. 2023-10-008). The need for informed consent was waived because of the retrospective nature of the study and the confidential record review [\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e].\u003c/p\u003e \u003c/div\u003e\n\u003ch3\u003eAI-based telemedicine platform for intracranial haemorrhage\u003c/h3\u003e\n\u003cp\u003eUnlike Seoul or metropolitan cities, general practitioners (GPs) or non-specialists in ICH oversee ERs in most rural areas in Gangwon province. Doctors lacking ICH experience are inevitably unable to perform swift diagnoses and provide timely treatment during the acute period, which potentially leads to unfavourable neurological outcomes. To increase the treatment quality of ICH patients who visit local hospitals in rural and medically underserved areas, we initiated the development of a cloud-based platform in collaboration with the government and information technology companies in 2022 (Project Number: 2710000241, RS-2022-00155659) [\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e, \u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e]. This platform was developed to strengthen the clinical capabilities of doctors with less ICH expertise by providing AI assistance with CT interpretation and teleconsultation (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eB) [\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e, \u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e]. To be more specific, a doctor at a local hospital in a rural area uploads whole CT images to the cloud, and then AI automatically determines whether a haemorrhage is present. Based on this interpretation, the doctor can request teleconsultation regarding the acute management of ICH and interhospital transfer. In addition, an ICH expert at the hub hospital can provide initial guidance on blood pressure control and securing an airway, and if a webcam is installed, neurological experts can deliver real-time remote consultation via video. Thus, this system facilitates the rapid transfer of patients requiring surgical procedures. In summary, the AI and telemedicine platform provides timely ICH diagnosis and enables proper initial treatment for spontaneous ICH and TBI and rapid transfer to a regional hub hospital for critical care and surgical treatment. In addition, the telemedicine platform can help doctors and nurses without ICH expertise in real settings.\u003c/p\u003e\n\u003ch3\u003eSocial media-based patient transfer in acute stroke\u003c/h3\u003e\n\u003cp\u003eThe objective of this project was to establish a network between regional hub hospitals and nearly local hospitals that cannot treat acute stroke. KakaoTalk, the most used messenger service in Korea (launched by Kakao Co. Ltd. in 2010), was used to construct the platform [\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e]. Briefly, the platform is operated as follows: 1) A doctor referring a stroke patient enters the KakaoTalk channel and selects from the menu a transfer request; 2) Basic patient information, vital signs, and key CT or MR images are entered; 3) A notification message is then sent to medical professionals registered on the channel; 4) Final transfer destination is determined based on considerations of recipient hospital intensive care unit and surgical availabilities. The main difference between the devised AI-based telemedicine platform and a previously reported AI-based telemedicine platform is that it is designed for patients with a diagnosis of stroke. Therefore, participants are neurosurgeons, neurologists, or emergency medicine doctors, not general doctors or nurses. The platform focuses on facilitating the transfer of stroke patients eligible for endovascular treatments or surgical procedures. The project is being conducted mainly in the cities of Chuncheon and Wonju.\u003c/p\u003e"},{"header":"Results","content":"\u003cdiv id=\"Sec7\" class=\"Section2\"\u003e \u003ch2\u003eAI-based telemedicine platform\u003c/h2\u003e \u003cp\u003eEight patients used the AI-based platform for diagnosis and treatment. Mean patient age was 70.5 years (range: 57\u0026ndash;87), and five were men (Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e). Most were diagnosed with TBI, for example, subdural or epidural haematoma. The remainder were suspected of having ICH initially, but there was no evidence of haemorrhage requiring further treatment. Senen patients were managed conservatively, and one underwent surgery. Four of the five patients transferred to a hub hospital received intensive care unit treatment. Their mean ICU stay was 2.3 days, and their hospital stays ranged from 4 and 14 days. All patients had favourable outcomes at 3 months. GPs referred six of the patients, and a nurse and an internist referred one patient apiece.\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab1\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 1\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eBaseline characteristics and clinical features of AI-based telemedicine and the social media-based platform for patients with acute brain conditions who present at rural hospitals\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"10\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c7\" colnum=\"7\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c8\" colnum=\"8\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c9\" colnum=\"9\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c10\" colnum=\"10\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eType\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eNo\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eAge/sex\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eDiagnosis\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003eReferral medical staff\u003c/p\u003e \u003cp\u003ein rural hospitals\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c6\"\u003e \u003cp\u003eMedical staff\u003c/p\u003e \u003cp\u003ein a hub hospital\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c7\"\u003e \u003cp\u003eGCS\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c8\"\u003e \u003cp\u003eManagement\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c9\"\u003e \u003cp\u003eICU stay\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c10\"\u003e \u003cp\u003e3-month\u003c/p\u003e \u003cp\u003emRS\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\" morerows=\"7\" rowspan=\"8\"\u003e \u003cp\u003eAI-based telemedicine\u003c/p\u003e \u003cp\u003eplatform\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e73/M\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eSeizure\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eNurse\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003eㅡ\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e15\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003eConservative\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e \u003cp\u003eㅡ\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c10\"\u003e \u003cp\u003e0\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e2\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e63/M\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eTBI\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eGeneral practitioners\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003eNeurosurgeon\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e15\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003eSurgery\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e \u003cp\u003e2\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c10\"\u003e \u003cp\u003e0\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e3\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e78/M\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eTBI\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eGeneral practitioners\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003eNeurosurgeon\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e15\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003eConservative\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e \u003cp\u003e0\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c10\"\u003e \u003cp\u003e0\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e4\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e57/M\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eNormal finding\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eGeneral practitioners\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003eㅡ\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e15\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003eConservative\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e \u003cp\u003eㅡ\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c10\"\u003e \u003cp\u003e0\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e5\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e72/F\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eTBI\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eGeneral practitioners\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003eNeurosurgeon\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e15\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003eConservative\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e \u003cp\u003e2\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c10\"\u003e \u003cp\u003e0\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e6\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e87/F\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eTBI\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eInternist\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003eNeurosurgeon\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e15\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003eConservative\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e \u003cp\u003e7\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c10\"\u003e \u003cp\u003e0\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e7\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e62/F\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eNormal finding\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eGeneral practitioners\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003eㅡ\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e15\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003eConservative\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e \u003cp\u003eㅡ\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c10\"\u003e \u003cp\u003e0\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e8\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e72/M\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eTBI\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eNurse\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003eNeurosurgeon\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e15\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003eConservative\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e \u003cp\u003e3\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c10\"\u003e \u003cp\u003e0\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\" morerows=\"10\" rowspan=\"11\"\u003e \u003cp\u003eSocial media-based\u003c/p\u003e \u003cp\u003eplatform\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e57/M\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eHaemorrhage stroke\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eNeurosurgeon\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003eNeurosurgeon\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e15\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003eCoil embolization\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e \u003cp\u003e14\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c10\"\u003e \u003cp\u003e0\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e2\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e63/M\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eIschemic stroke\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eNeurosurgeon\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003eNeurologist\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e15\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003eConservative\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e \u003cp\u003e0\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c10\"\u003e \u003cp\u003e0\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e3\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e64/M\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eTBI\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eNeurosurgeon\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003eNeurosurgeon\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e15\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003eConservative\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e \u003cp\u003e5\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c10\"\u003e \u003cp\u003e0\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e4\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e86/M\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eIschemic stroke\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eNeurosurgeon\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003eNeurologist\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e13\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003eConservative\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e \u003cp\u003e3\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c10\"\u003e \u003cp\u003e2\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e5\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e34/M\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eTBI\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eNeurosurgeon\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003eNeurosurgeon\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e15\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003eConservative\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e \u003cp\u003e8\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c10\"\u003e \u003cp\u003e0\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e6\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e95/M\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eNormal finding\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eNeurosurgeon\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003eㅡ\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e15\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003eㅡ\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e \u003cp\u003eㅡ\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c10\"\u003e \u003cp\u003e0\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e7\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e66/M\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eHaemorrhage stroke\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eNeurosurgeon\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003eNeurosurgeon\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e15\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003eConservative\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e \u003cp\u003e5\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c10\"\u003e \u003cp\u003e2\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e8\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e54/F\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eIschemic stroke\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eNeurosurgeon\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003eNeurosurgeon\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e15\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003eConservative\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e \u003cp\u003e3\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c10\"\u003e \u003cp\u003e1\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e9\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e64/M\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eIschemic stroke\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eEmergency physician\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003eNeurologist\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e12\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003eIA Thrombectomy\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e \u003cp\u003e23\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c10\"\u003e \u003cp\u003e6\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e10\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e56/M\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eHaemorrhage stroke\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eNeurosurgeon\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003eNeurosurgeon\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e5\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003eSurgery\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e \u003cp\u003e32\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c10\"\u003e \u003cp\u003e4\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e11\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e72/M\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eHaemorrhage stroke**\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eNeurosurgeon\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003eNeurosurgeon\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e3\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003eConservative\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e \u003cp\u003e0*\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c10\"\u003e \u003cp\u003e6\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e12\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e75/F\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eTBI\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eNeurosurgeon\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003eNeurosurgeon\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e15\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003eConservative\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e \u003cp\u003e21\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c10\"\u003e \u003cp\u003e1\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003ctfoot\u003e \u003ctr\u003e\u003ctd colspan=\"10\"\u003eAI, artificial intelligence; GCS, Glasgow Coma Scale; IA, intraarterial; ICU, intensive care unit; mRS, modified Rankin Scale; TBI; traumatic brain injury\u003c/td\u003e\u003c/tr\u003e \u003ctr\u003e\u003ctd colspan=\"10\"\u003eㅡ indicates that the patient was not hospitalized at a hub hospital. 0* indicates death in the emergency room of a hub hospital.\u003c/td\u003e\u003c/tr\u003e \u003c/tfoot\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec8\" class=\"Section2\"\u003e \u003ch2\u003eIllustrations of representative cases\u003c/h2\u003e \u003cp\u003eA 63-year-old male visited a local ER for an aggravated, unbearable headache with dizziness. He had been involved in a motorcycle accident a month previously. In the absence of a doctor with ICH expertise in the ER, the duty doctor uploaded CT images to the cloud. AI embedded in the platform then identified subdural haematoma on the left side and marked this using different colours (Figs.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eA and B). The patient was immediately transferred, with telemedicine consultation, to a regional hub hospital for surgery (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eC). A CT scan taken 3 months later revealed almost total elimination of the haematoma (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eD).\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eThe following case is an example of AI and teleconsultation use by an internist. An 86-year-old woman presented with progressive headache after a fall a week previously. AI identified and marked a haematoma at the site of the anterior falx (Figs.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eE and F). At this time, the internist consulted a neurosurgeon at the hub hospital in real time by teleconsultation regarding the initial treatment and transfer (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eE).\u003c/p\u003e \u003cp\u003eThe final example involves the detection of a small amount of subdural haemorrhage by AI in a male patient admitted to an orthopaedic unit for multiple traumas who complained of a persistent headache. He was taking anticoagulants for arterial fibrillation and valvular heart disease, and thus, the doctor was concerned about the possibility of brain haemorrhage. However, no specialist capable of diagnosing and treating cerebral haemorrhage was available. In this case, the AI platform ​​identified subdural haemorrhage (a white arrow, Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eG), which may not have been easily detected due to its proximity to bone. The patient was discharged after consulting a neurosurgeon.\u003c/p\u003e \u003c/div\u003e\n\u003ch3\u003eSocial media-based patient transfer\u003c/h3\u003e\n\u003cp\u003eTwenty-four patients used the patient transfer platform during the study period (six months, October 2024 \u0026ndash; March 2025), and the 12\u003c/p\u003e \u003cp\u003epatients admitted to a single institution in Chuncheon City were included. Mean patient age was 67.8 years, with ten men and two women. The diagnoses were ischemic stroke in 4 patients,\u003c/p\u003e \u003cp\u003ehaemorrhagic stroke in 4 patients, and TBI in 3 patients. One patient was referred for suspected cerebral infarction but discharged after a negative diffusion-weighted MRI finding. Eight patients were managed conservatively, and three patients underwent surgical procedures, including coil embolization, IA thrombectomy, or haematoma stereotactic catheter placement for haematoma aspiration (Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e). Mean ICU length of stay was 10.36 days. Of the \u003cb\u003eeleven\u003c/b\u003e survivors, seven achieved favourable functional outcomes.\u003c/p\u003e\n\u003ch3\u003eIllustrations of representative cases\u003c/h3\u003e\n\u003cp\u003eThe first representative case involved a patient diagnosed with haemorrhagic stroke in the ER of a university hospital who was transferred to another university hospital for emergent surgery. A 56-year-old male with hypertension and gout presented with a stuporous mentality. Initial CT findings showed multiple haemorrhagic lesions of thalamus, ventricles, and parietal lobe (black arrows, Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eA). However, the on-duty neurosurgeon was performing surgery, and thus, an emergent procedure could not be performed immediately. The neurosurgeon inquired through the KakaoTalk platform whether the patient could be operated on immediately at another hospital, and within 5 minutes, a hospital was identified that could perform the surgery immediately. Two catheters targeting the ventricle and haematoma were placed to increase ICP control by cerebrospinal fluid and haematoma drainage. A CT scan taken three days later revealed the haematoma had decreased, which concurred with relieved ICP (Figs.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eB and C). Two-month follow-up CT showed the haematoma had disappeared without ventricular enlargement (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eD). At this time, the patient exhibited near-normal levels of consciousness and was transferred to a rehabilitation hospital.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eThe second case involves the timely transfer of a patient with acute cerebral infarction to a hub hospital providing neurocritical care. An 86-year-old man visited a local ER for right-side motor weakness with diplopia of two days duration. Diffusion MRI revealed acute cerebral infarctions in the vascular territory of the posterior cerebral artery (white arrows, Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eE). Within 5 minutes of placing a transfer request on the platform, the patient was matched to a hospital capable of neurocritical care. One month later, the decline in motor performance of right extremities had improved to grade IV\u0026thinsp;+\u0026thinsp;from IV at the time of admission, and the infarction had not increased in extent (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eF).\u003c/p\u003e \u003cp\u003eThe third case involved the transfer of a hemodynamically unstable patient. A 72-year-old male was found unconscious in a parking lot. He was immediately transferred to the ER of a nearby rural hospital and was diagnosed with severe brain swelling with multiple cortical SAH, SDH, and fractures of the right temporal and occipital bones with associated lesions (white arrows, Figs.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eG and H). CPR was performed due to sudden cardiac arrest, and a request for patient transfer was made via KakaoTalk for professional intensive care management. The patient was transferred to a hub hospital without delay. Nevertheless, the patient died despite continuous CPR after transport. This case demonstrates that hemodynamically unstable patients should receive continuous monitoring and treatment by specialists during transport.\u003c/p\u003e"},{"header":"Discussion","content":"\u003cp\u003eKorea has made remarkable progress in the fields of economics, technology, and medicine over the last three decades. However, this rapid growth inevitably has led to differences in the distribution of regional wealth and resulted in qualitative and quantitative differences in healthcare systems [\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e]. Although regional differences in overall avoidable and preventable deaths in metropolitan and non-metropolitan areas were alleviated somewhat between 1995 and 2019 [\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e], the treatment of acute brain conditions is still subject to regional disparities [\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e]. Furthermore, under the conditions imposed by the COVID-19 pandemic, it became difficult to treat critically neurologically impaired patients. A nationwide web- and mobile phone-based teleconsultation network for rural and urban hospitals has been established in South Korea specifically for the treatment of emergent patients [\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e]. The remote emergency consultation afforded by this system achieved reductions in the unnecessary transportation of trauma patients without severe injuries [\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e]. Lee et al. [\u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e] also reported that video meetings with caregivers using a mobile device increase caregiver satisfaction under the restrictions imposed due to COVID-19. However, the nationwide emergency teleconsultation network is more focused on remote imaging interpretation of overall general emergencies or the transfer of trauma patients than on patients with acute brain conditions such as stroke [\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e]. Furthermore, mobile phone-based video interviews for neurosurgical critically ill patients are not designed to strengthen the clinical capacity of rural hospitals to treat the acute phase of acute brain conditions [\u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e]. The \u0026ldquo;golden hour\u0026rdquo; is an important treatment maxim for acute brain conditions, which require two completely different surgical procedures, viz., craniotomy through skull opening or endovascular treatment via femoral and radial arteries. Thus, it is important to nominate hub hospitals that can provide appropriate treatment, especially in rural areas where treatment is not possible. In addition, accurate diagnoses by medical imaging and the provision of early, timely treatment by doctors who encounter neurologically ill patients for the first time in rural ERs are often overlooked priorities. CT and MRI (magnetic resonance imaging) are primarily used to identify brain abnormalities. However, though most ERs in rural Korea have CT units, relatively few have MRI instruments. Consequently, when developing an AI-based automatic interpreting application, CT provides the best means of enhancing the diagnostic capabilities of rural ERs.\u003c/p\u003e \u003cp\u003eEarly medical treatment, like surgical or endovascular intervention, is also essential for achieving favourable neurological outcomes. In particular, the amount of haemorrhage haematoma increases within the first 6\u0026ndash;24 hours after ictus [\u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e, \u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e, \u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e]. Arima et al [\u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e]. reported that intensive blood pressure (BP) lowering achieved benefits in patients with a neurological condition, notably, to between a systolic BP of 130\u0026ndash;139 mm Hg in patients with acute ICH. In addition to BP control, appropriate airway, antiepileptic drug use, and increased intracranial pressure (IICP) management should be performed simultaneously when ICH is diagnosed, as these favourably influence neurological outcomes [\u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e, \u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e23\u003c/span\u003e]. However, the reality is that most doctors working in rural ERs in Korea are GPs, and even among specialist doctors, few are experts in acute brain conditions. Thus, swift diagnosis and timely appropriate treatment cannot be conducted effectively.\u003c/p\u003e \u003cp\u003eDigital health technologies are increasingly being used to transform care paradigms for patients with acute brain conditions [\u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e24\u003c/span\u003e]. In particular, the appropriate application of AI, mobile and wireless technologies, and cloud platform services could potentially improve treatment capabilities in rural hospitals through telemedicine. In the United States, so-called telestroke services are being actively utilized to diagnose ischemic stroke patients indicated for thrombolysis or endovascular thrombectomy rapidly. Zachrison et al [\u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e25\u003c/span\u003e]. reported that consultation start time and thrombolysis performance time decreased over time after introducing a telestroke platform. Also, telestroke significantly increased the rate of receiving treatment within 3 hours of symptom onset (OR\u0026thinsp;=\u0026thinsp;2.15; 95% CI 1.37\u0026ndash;3.40), improved neurological outcomes at 3 months (OR\u0026thinsp;=\u0026thinsp;1.29; 95% CI 1.01\u0026ndash;1.63), and lowered the in-hospital mortality rate (OR\u0026thinsp;=\u0026thinsp;0.67; 95% CI 0.52\u0026ndash;0.87) [\u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e26\u003c/span\u003e]. Ford et al. [\u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e27\u003c/span\u003e] revealed that telemedicine assessment reduced the times required to achieve BP control and anticoagulation reversal. Accordingly, the application of digital technologies in rural areas is likely associated with better prognoses in patients with an acute brain condition [\u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e26\u003c/span\u003e]. However, telemedicine is still not generally utilized due to regulations regarding patients with acute brain conditions, although the Korean government, prompted by the COVID-19 pandemic, temporarily allowed telemedicine [\u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e]. Fortunately, Gangwon-do has been designated a regulatory-free zone by the Korean government, which has made telemedicine possible.\u003c/p\u003e \u003cp\u003eHere, we implemented two platform types to address the special characteristics of Gangwon-do province. To design these platforms, we focused on the expertise of initial care physicians in rural hospitals and their requirements from neurological specialists in hub hospitals. For example, if a doctor does not have specialized knowledge about neurological abnormalities, AI could be used to aid the diagnosis and provide details of initial medical requirements such as BP control, respiration stabilization, and IICP management before transfer. Inexperienced doctors often have difficulty differentiating normal and abnormal conditions, and in such cases, AI interpretation with subsequent teleconsultation can aid diagnosis prior to transfer. Concern remains that the rapid pace of AI advancement may outpace the infrastructure capabilities of rural regions and pose a barrier to sustainable implementation. This issue can be addressed through a cloud-based platform, which offers a flexible environment for the implementation of artificial intelligence. On the other hand, when referring physicians at rural hospitals are neurosurgeons, neurologists, or emergency medicine specialists capable of diagnosing brain diseases by CT or MRI and initiating early medical treatment, the main issue is to transfer the patient to a hub as quickly as possible for surgical treatment if needed. Social media-based platforms enable effective communication between experts, and a social network approach enhances accessibility by leveraging a mobile communication platform used by over 95% of smartphone users in Korea. Nevertheless, further research is needed to determine whether large amounts of medical data can be processed reliably and stably in emergent situations.\u003c/p\u003e \u003cp\u003eWhile operating the artificial intelligence (AI)-based telemedicine and social media-based patient transfer platforms, we identified two issues that need to be addressed to maximize their effectiveness in medical settings. The first issue is to improve the effectiveness of cooperation, particularly for unstable patients with acute, severe brain conditions. The majority of cases subjected to digital technologies and transferred to a hub hospital did not have severe or unstable neurological complications. However, even when digital technology was applied, patients with acute and severe brain disease were transferred to a hub hospital from areas served by rural hospitals. We suppose such transfers are due to a fear of being presented with seriously ill patients. This fear can only be resolved by education, and we suggest virtual reality (VR) may be suitable in emergent and unstable situations. From the perspective of the learner, VR conveys complex issues quickly and has the potential to replace mannequin-based courses [\u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e28\u003c/span\u003e]. Accordingly, additional educational programs on severe neurological disorders should be provided to staff in rural ERs to ensure proper neurological examinations, diagnosis, and transportation. The second issue is integration with an AI platform capable of diagnosing ischemic stroke. According to Korean statistics, ischemic stroke accounts for about 70% of all stroke cases [\u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e29\u003c/span\u003e]. To facilitate the use of tissue-type plasminogen activator (tPA) in rural ERs, a CT-based AI algorithm to identify ischemic stroke is needed. Moreover, for effective management in rural ERs, it is necessary not only to have a highly accurate diagnostic algorithm for ischemic stroke, but also to implement functions that enable integration with neuromonitoring and prognosis estimation systems [\u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e30\u003c/span\u003e].\u003c/p\u003e \u003cp\u003eIn this study, we applied two different digital technologies in real-world clinical settings to improve rural healthcare environments and evaluated their impacts on the quality of treatment afforded to acute brain conditions. Through the indirect integration of the two platforms, we gained insights into how a more effective telemedicine system could be structured (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003e). Integrating different platforms allows individual telemedicine systems to be utilized beyond their intended purposes. However, such integration requires careful consideration to mitigate unforeseen risks associated with unintended use. The application of the most recent medical AI systems provides a means of addressing this challenge. The integration of medical AI and telemedicine can provide AI-assistant tools that help correct diagnostic errors by non-experts and generative AI systems. Recent studies have reported that AI-powered diagnostic support tools demonstrated excellent performance when assisting non-task-specific expert clinicians or non-experienced healthcare providers [\u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e31\u003c/span\u003e, \u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e32\u003c/span\u003e]. At present, generative AI requires the establishment of safety regulations and standardization to ensure reliability, but it has substantial potential for broader applications [\u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e33\u003c/span\u003e]. In particular, large language models (LLMs) offer better interpretations of complex medical terminology and summaries of large volumes of clinical information [\u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e34\u003c/span\u003e, \u003cspan citationid=\"CR35\" class=\"CitationRef\"\u003e35\u003c/span\u003e]. In addition to providing early diagnosis, LLMs offer comprehensive and detailed guidance on treatment options and use hospital-specific terminology or contextual factors in the guidance process. By leveraging relatively flexible data structures, generative AI enables the expansion of telemedicine platforms into home-based rehabilitation through technologies such as the Internet of Things (IoT) and is anticipated to play a pivotal role in the future development of telemedicine.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e"},{"header":"Conclusions","content":"\u003cp\u003eAI-based telemedicine and social media-based transfer platforms offer a promising solution to the practical challenges posed by limited medical infrastructure in rural areas. Their implementation facilitates early diagnosis and appropriate treatment for patients with acute brain conditions and provides efficient interhospital transfers regardless of the level of the initial hospital. These findings suggest that long-term, innovative, digital health projects are essential for securing treatment enhancements in the future rural medical environment.\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eEthics approval and consent to participate\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThis study was approved by the institutional review board of Hallym University Chuncheon Sacred Heart Hospital (Approval number. IRB 2021-10-012-007). All experimental methods complied with the Helsinki Declaration. The need for informed consent was waived by the IRB of our hospital due to the retrospective design of the study.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eConsent for publication\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eNot applicable.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eData availability\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003ePlease direct enquiries to the corresponding author.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eConflicts of interest\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe authors have no conflict of interest to declare.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eFunding\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThis work was supported by a Korea Medical Device Development Fund grant funded by the Ministry of Science and ICT, the Ministry of Trade, Industry and Energy, the Ministry of Health \u0026amp; Welfare, the Ministry of Food and Drug Safety (Project Number: 2710000241, RS-2022-00155659), the Pilot Project for Severe and Emergency Cardiovascular Disease Problem-Solving Treatment Cooperation Network, the Hallym University Medical Center Research Fund, and the Hallym University Research Fund (HURF).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAuthors\u0026apos; Contributions\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eMSP and S-HS contributed to data curation, formal analysis, and drafting of the original manuscript. JPJ and CK were involved in conceptualization, methodology development, formal analysis, investigation, funding acquisition, and manuscript drafting. SH, JK, S-HL, J-KR, SJ, SHS, JEK, J-HS, HJC, HSJ, JHA, SJL, SK, JJL, YHI, HJ, HK, JSY, SHK, and YJC contributed to data collection, data validation, and resource provision. All authors reviewed and approved the final version of the manuscript for submission.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\u003cli\u003e\u003cspan\u003eJeon JP, Lee SU, Kim SE, Kang SH, Yang JS, Choi HJ et al. Correlation of optic nerve sheath diameter with directly measured intracranial pressure in Korean adults using bedside ultrasonography. PLoS One. 2017;12(9):e0183170. PMID: 28902893. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1371/journal.pone.0183170\u003c/span\u003e\u003cspan address=\"10.1371/journal.pone.0183170\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eVespa P. Continuous EEG monitoring for the detection of seizures in traumatic brain injury, infarction, and intracerebral hemorrhage: to detect and protect. J Clin Neurophysiol. 2005;22(2):99\u0026ndash;106. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1097/01.wnp.0000154919.54202.e0\u003c/span\u003e\u003cspan address=\"10.1097/01.wnp.0000154919.54202.e0\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e. PMID: 15805809.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePark YK, Yoon BH, Won YD, Kim JH, Kang HI. Real-World Impact of Modern Reperfusion Therapy for Acute Ischemic Stroke: A Nationwide Population-Based Data Study in Korea. J Korean Neurosurg Soc. 2024;67(2):186\u0026ndash;93. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.3340/jkns.2023.0133\u003c/span\u003e\u003cspan address=\"10.3340/jkns.2023.0133\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e. PMID: 37799025.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePark HA, Vaca FE, Jung-Choi K, Park H, Park JO. Area-Level Socioeconomic Inequalities in Intracranial Injury-Related Hospitalization. J Korean Med Sci. 2023;38(4):e38. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.3346/jkms.2023.38.e38\u003c/span\u003e\u003cspan address=\"10.3346/jkms.2023.38.e38\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e. PMID: 36718564. in Korea: A Retrospective Analysis of Data From Korea National Hospital Discharge Survey 2008\u0026ndash;2015.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePark HK, Kim SE, Cho YJ, Kim JY, Oh H, Kim BJ, et al. Quality of acute stroke care in Korea (2008\u0026ndash;2014): Retrospective analysis of the nationwide and nonselective data for quality of acute stroke care. Eur Stroke J. 2019;4(4):337\u0026ndash;46. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1177/2396987319849983\u003c/span\u003e\u003cspan address=\"10.1177/2396987319849983\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e. PMID: 31903432.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKorea S, e-Nara I. 2025 March 24;\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://www.index.go.kr/unity/potal/main/EachDtlPageDetail.do?idx_cd=1007\u003c/span\u003e\u003cspan address=\"https://www.index.go.kr/unity/potal/main/EachDtlPageDetail.do?idx_cd=1007\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKim YJ, Li L, Hwang JY. A Maternity Waiting Home Is an Alternative Approach for the Accessibility of Pregnant Women in an Obstetrically Underserved Area of Korea. J Korean Med Sci. 2023;38(17):e164. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.3346/jkms.2023.38.e164\u003c/span\u003e\u003cspan address=\"10.3346/jkms.2023.38.e164\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e. PMID: 37128881.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKim JY, Kang K, Kang J, Koo J, Kim DH, Kim BJ, et al. Executive Summary of Stroke Statistics in Korea 2018: A Report from the Epidemiology Research Council of the Korean Stroke Society. J Stroke. 2019;21(1):42\u0026ndash;59. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.5853/jos.2018.03125\u003c/span\u003e\u003cspan address=\"10.5853/jos.2018.03125\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e. PMID: 30558400.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKim JT, Fonarow GC, Smith EE, Reeves MJ, Navalkele DD, Grotta JC, et al. Treatment With Tissue Plasminogen Activator in the Golden Hour and the Shape of the 4.5-Hour Time-Benefit Curve in the National United States Get With The Guidelines-Stroke Population. Circulation. 2017;135(2):128\u0026ndash;39. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1161/CIRCULATIONAHA.116.023336\u003c/span\u003e\u003cspan address=\"10.1161/CIRCULATIONAHA.116.023336\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e. PMID: 27815374.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRandhawa AS, Pariona-Vargas F, Starkman S, Sanossian N, Liebeskind DS, Avila G, et al. Beyond the Golden Hour: Treating Acute Stroke in the Platinum 30 Minutes. Stroke. 2022;53(8):2426\u0026ndash;34. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1161/STROKEAHA.121.036993\u003c/span\u003e\u003cspan address=\"10.1161/STROKEAHA.121.036993\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e. PMID: 35545939.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMisirlioglu M, Ekinci F, Yildizdas D, Horoz OO, Yilmaz HL, Incecik F, et al. A Retrospective Cohort Study of Traumatic Brain Injury in Children: A Single-Institution Experience and Determinants of Neurologic Outcome. J Crit Care Med (Targu Mures). 2023;9(4):252\u0026ndash;61. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.2478/jccm-2023-0027\u003c/span\u003e\u003cspan address=\"10.2478/jccm-2023-0027\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e. PMID: 37969881.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eJun HS, Yang K, Kim J, Jeon JP, Ahn JH, Lee SJ, et al. Development of Cloud-Based Telemedicine Platform for Acute Intracerebral Hemorrhage in Gangwon-do: Concept and Protocol. J Korean Neurosurg Soc. 2023;66(5):488\u0026ndash;93. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.3340/jkns.2022.0256\u003c/span\u003e\u003cspan address=\"10.3340/jkns.2022.0256\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e. PMID: 36756670.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eJun HS, Yang K, Kim J, Jeon JP, Kim SJ, Ahn JH, et al. Telemedicine Protocols for the Management of Patients with Acute Spontaneous Intracerebral Hemorrhage in Rural and Medically Underserved Areas in Gangwon State: Recommendations for Doctors with Less Expertise at Local Emergency Rooms. J Korean Neurosurg Soc. 2024;67(4):385\u0026ndash;96. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.3340/jkns.2023.0199\u003c/span\u003e\u003cspan address=\"10.3340/jkns.2023.0199\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e. PMID: 37901932.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eYun TJ, Choi JW, Han M, Jung WS, Choi SH, Yoo RE et al. Deep learning based automatic detection algorithm for acute intracranial haemorrhage: a pivotal randomized clinical trial. NPJ Digit Med. 2023;6(1):61. PMID: 37029272. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1038/s41746-023-00798-8\u003c/span\u003e\u003cspan address=\"10.1038/s41746-023-00798-8\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWoo SH, Oh EG, Kim KS, Chu SH, Kim GS, Nam CM. Development and Assessment of a Social Network Service-Based Lifestyle-Modification Program for Workers at High Risk of Developing Cardiovascular Disease. Workplace Health Saf. 2020;68(3):109\u0026ndash;20. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1177/2165079919864976\u003c/span\u003e\u003cspan address=\"10.1177/2165079919864976\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e. PMID: 31434552.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eJung EJ, Kim DY, Bae HJ, Ko KP. Assessing regional disparities and vulnerability in stroke care across Gyeonggi Province: A focus on hospital service areas. J Stroke Cerebrovasc Dis. 2024;33(9):107817. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1016/j.jstrokecerebrovasdis.2024.107817\u003c/span\u003e\u003cspan address=\"10.1016/j.jstrokecerebrovasdis.2024.107817\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e. PMID: 38880365.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eChoi MH, Moon MH, Yoon TH. Avoidable Mortality between Metropolitan and Non-Metropolitan Areas in Korea from 1995 to 2019: A Descriptive Study of Implications for the National Healthcare Policy. Int J Environ Res Public Health. 2022;19(6). PMID: 35329162. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.3390/ijerph19063475\u003c/span\u003e\u003cspan address=\"10.3390/ijerph19063475\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eChoi W, Lim Y, Heo T, Lee S, Kim W, Kim SC, et al. Characteristics and Effectiveness of Mobile- and Web-Based Tele-Emergency Consultation System between Rural and Urban Hospitals in South Korea: A National-Wide Observation Study. J Clin Med. 2023;12(19). \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.3390/jcm12196252\u003c/span\u003e\u003cspan address=\"10.3390/jcm12196252\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e. PMID: 37834896.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLee MH, Jang SR, Lee TK. The Direction of Neurosurgery to Overcome the Living with COVID-19 Era: The Possibility of Telemedicine in Neurosurgery. J Korean Neurosurg Soc. 2023;66(5):573\u0026ndash;81. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.3340/jkns.2022.0211\u003c/span\u003e\u003cspan address=\"10.3340/jkns.2022.0211\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e. PMID: 37667635.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBrouwers HB, Greenberg SM. Hematoma expansion following acute intracerebral hemorrhage. Cerebrovasc Dis. 2013;35(3):195\u0026ndash;201. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1159/000346599\u003c/span\u003e\u003cspan address=\"10.1159/000346599\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e. PMID: 23466430.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKuohn LR, Witsch J, Steiner T, Sheth KN, Kamel H, Navi BB, et al. Early Deterioration, Hematoma Expansion, and Outcomes in Deep Versus Lobar Intracerebral Hemorrhage: The FAST Trial. Stroke. 2022;53(8):2441\u0026ndash;8. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1161/STROKEAHA.121.037974\u003c/span\u003e\u003cspan address=\"10.1161/STROKEAHA.121.037974\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e. PMID: 35360929.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eArima H, Heeley E, Delcourt C, Hirakawa Y, Wang X, Woodward M, et al. Optimal achieved blood pressure in acute intracerebral hemorrhage: INTERACT2. Neurology. 2015;84(5):464\u0026ndash;71. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1212/WNL.0000000000001205\u003c/span\u003e\u003cspan address=\"10.1212/WNL.0000000000001205\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e. PMID: 25552575.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLi Z, You M, Long C, Bi R, Xu H, He Q et al. Hematoma Expansion in Intracerebral Hemorrhage: An Update on Prediction and Treatment. Front Neurol. 2020;11:702. PMID: 32765408. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.3389/fneur.2020.00702\u003c/span\u003e\u003cspan address=\"10.3389/fneur.2020.00702\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSilva GS, Andrade JBC. Digital health in stroke: a narrative review. Arq Neuropsiquiatr. 2024;82(8):1\u0026ndash;10. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1055/s-0044-1789201\u003c/span\u003e\u003cspan address=\"10.1055/s-0044-1789201\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e. PMID: 39187259.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZachrison KS, Sharma R, Wang Y, Mehrotra A, Schwamm LH. National Trends in Telestroke Utilization in a US Commercial Platform Prior to the COVID-19 Pandemic. J Stroke Cerebrovasc Dis. 2021;30(10):106035. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1016/j.jstrokecerebrovasdis.2021.106035\u003c/span\u003e\u003cspan address=\"10.1016/j.jstrokecerebrovasdis.2021.106035\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e. PMID: 34419836.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLazarus G, Permana AP, Nugroho SW, Audrey J, Wijaya DN, Widyahening IS. Telestroke strategies to enhance acute stroke management in rural settings: A systematic review and meta-analysis. Brain Behav. 2020;10(10):e01787. PMID: 32812380. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1002/brb3.1787\u003c/span\u003e\u003cspan address=\"10.1002/brb3.1787\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eFord S, Ajani Z, Chen Q, Sorreda V, Tu G, McCartney D, et al. Comparison of Standard Emergency Room Care with Tele-Stroke Evaluation in Acute Intracerebral Hemorrhage Management (P6. 030). Neurology. 2016;86(16supplement):P6.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMahling M, Wunderlich R, Steiner D, Gorgati E, Festl-Wietek T, Herrmann-Werner A. Virtual Reality for Emergency Medicine Training in Medical School: Prospective, Large-Cohort Implementation Study. J Med Internet Res. 2023;25:e43649. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.2196/43649\u003c/span\u003e\u003cspan address=\"10.2196/43649\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e. PMID: 36867440.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHong KS, Bang OY, Kim JS, Heo JH, Yu KH, Bae HJ, et al. Stroke Statistics in Korea: Part II Stroke Awareness and Acute Stroke Care, A Report from the Korean Stroke Society and Clinical Research Center For Stroke. J Stroke. 2013;15(2):67\u0026ndash;77. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.5853/jos.2013.15.2.67\u003c/span\u003e\u003cspan address=\"10.5853/jos.2013.15.2.67\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e. PMID: 24324942.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eUparela-Reyes MJ, Villegas-Trujillo LM, Cespedes J, Velasquez-Vera M, Rubiano AM. Usefulness of Artificial Intelligence in Traumatic Brain Injury: A Bibliometric Analysis and Mini-review. World Neurosurg. 2024;188:83\u0026ndash;92. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1016/j.wneu.2024.05.065\u003c/span\u003e\u003cspan address=\"10.1016/j.wneu.2024.05.065\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e. PMID: 38759786.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eYanagawa M. Artificial Intelligence Improves Radiologist Performance for Predicting Malignancy at Chest CT. Radiology. 2022;304(3):692\u0026ndash;3. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1148/radiol.220571\u003c/span\u003e\u003cspan address=\"10.1148/radiol.220571\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e. PMID: 35608448.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGaube S, Suresh H, Raue M, Lermer E, Koch TK, Hudecek MFC et al. Non-task expert physicians benefit from correct explainable AI advice when reviewing X-rays. Sci Rep. 2023;13(1):1383. PMID: 36697450. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1038/s41598-023-28633-w\u003c/span\u003e\u003cspan address=\"10.1038/s41598-023-28633-w\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGoodman KE, Yi PH, Morgan DJ. AI-Generated Clinical Summaries Require More Than Accuracy. JAMA. 2024;331(8):637\u0026ndash;8. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1001/jama.2024.0555\u003c/span\u003e\u003cspan address=\"10.1001/jama.2024.0555\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e. PMID: 38285439.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eShool S, Adimi S, Saboori Amleshi R, Bitaraf E, Golpira R, Tara M. A systematic review of large language model (LLM) evaluations in clinical medicine. BMC Med Inform Decis Mak. 2025;25(1):117. PMID: 40055694. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1186/s12911-025-02954-4\u003c/span\u003e\u003cspan address=\"10.1186/s12911-025-02954-4\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWals Zurita AJ, Miras Del Rio H, Ugarte Ruiz de Aguirre N, Nebrera Navarro C, Rubio Jimenez M, Munoz Carmona D, et al. The Transformative Potential of Large Language Models in Mining Electronic Health Records Data: Content Analysis. JMIR Med Inf. 2025;13:e58457. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.2196/58457\u003c/span\u003e\u003cspan address=\"10.2196/58457\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e. PMID: 39746191.\u003c/span\u003e\u003c/li\u003e\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":true,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true},"keywords":"Telemedicine, Artificial intelligence, Social media, Health inequity, Digital technology","lastPublishedDoi":"10.21203/rs.3.rs-6542673/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-6542673/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003ch2\u003eBackground\u003c/h2\u003e \u003cp\u003eDespite nationwide efforts to enhance the quality of treatment for acute brain conditions in Korea, regional disparities persist due to the lack of neurology specialists and infrastructure shortcomings in rural areas.\u003c/p\u003e\u003ch2\u003eMethods\u003c/h2\u003e \u003cp\u003e We implemented two digital technologies, namely, artificial intelligence (AI)-based telemedicine and social media-based patient transfer platforms, from January 2024 to improve treatment quality for early-stage patients with various brain conditions in rural hospitals and facilitate links with regional hub hospitals. Here, we review medical records, share our experience of using digital technologies, and address current limitations and future perspectives.\u003c/p\u003e\u003ch2\u003eResults\u003c/h2\u003e \u003cp\u003eThe AI-based platform was installed to facilitate collaboration between non-experts at rural hospitals and experts at hub hospitals, and the social media-based platform was adopted to improve collaboration between experts. Eight patients with a mean age of 70.7 years used the AI-based platform to facilitate accurate diagnosis and treatment. The non-experts who referred patients included general practitioners (n\u0026thinsp;=\u0026thinsp;5, 62.5%), an internist (n\u0026thinsp;=\u0026thinsp;1, 12.5%), and nurses (n\u0026thinsp;=\u0026thinsp;2, 25.0%). The platform enabled rapid diagnosis and decision-making, and its use led to favourable outcomes. The social media-based platform was used to transfer 12 diagnosed patients. Eleven patients (91.7%) received neurocritical care, and three (25.0%) underwent surgical procedures at a hub hospital after transfer. Nine patients (75.0%) had favourable outcomes.\u003c/p\u003e\u003ch2\u003eConclusion\u003c/h2\u003e \u003cp\u003eWe suggest a novel means of reducing regional inequities in the treatment of acute brain conditions that addresses the diversity of rural medical environments. The two digital technologies implemented have helped rural hospitals respond early and facilitated inter-hospital transfer. Additional features that consider user convenience and automatic linkage of diagnosis and treatment are essential to enable the nationwide expansion of the above platforms.\u003c/p\u003e","manuscriptTitle":"Artificial Intelligence and Social Media Utilization for Rural Patients with Acute Brain Conditions in Chuncheon, Gangwon-do, South Korea","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-05-22 11:03:04","doi":"10.21203/rs.3.rs-6542673/v1","editorialEvents":[{"type":"communityComments","content":0}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"20d19376-2ece-45d1-b7e5-1cac3864c525","owner":[],"postedDate":"May 22nd, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[],"tags":[],"updatedAt":"2026-01-21T11:36:18+00:00","versionOfRecord":[],"versionCreatedAt":"2025-05-22 11:03:04","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-6542673","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-6542673","identity":"rs-6542673","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}