Utilization of AI by the doctors for patient care in tertiary care hospital Peshawar

Utilization of AI by the doctors for patient care in tertiary care hospital Peshawar | 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 Utilization of AI by the doctors for patient care in tertiary care hospital Peshawar Amna Faiz, Rizwan Ul Haq, Anees Ur Rehman, Hammad Ullah, Quratulain Zahir, and 1 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-7290827/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 Introduction: Artificial intelligence is strengthening modern healthcare by assisting with diagnosis, increasing patient monitoring, and lowering physician burden. Despite the fact that industrialized nations have begun to incorporate it into their healthcare systems, developing nations such as Pakistan continue to lag behind. Objective: This study assesses physicians' opinions and understanding about using AI in patient care at Peshawar's tertiary care institutions. Methods: This descriptive cross-sectional study was carried out between April and August of 2024 using a self-administered, closed-ended, structured questionnaire. A total of 355 people from a range of professional backgrounds participated in the study. According to the results, just 40.6% of the doctors actually employed AI, despite 91.1% of them being aware of its usage. Results: The results of an exploratory factor analysis showed a moderate but statistically significant correlation between AI knowledge and use (r=0.15, p=0.013). The data's eligibility for factor analysis was validated by the Bartlett's Test of Sphericity (p = 0.006) and the significant Kaiser-Meyer-Olkin score (0.500). Fisher's Exact test emphasized the importance of knowledge and implementation. Conclusion: Despite widespread knowledge, there is little actual use of AI due to obstacles including infrastructure, support, and training. These results demonstrate how urgently Pakistan needs organized AI infrastructure and training to support the growth of its healthcare sector. Improving doctors' proficiency and trust in AI might close the knowledge gap and lead to better patient care. Introduction Artificial intelligence (AI) has drawn extraordinary interest in recent decades and is being referred to be the fourth industrial revolution.[ 1 ] There are several definitions of AI, however, at the moment, Boden's definition—the capacity to enable computers to perform tasks that would need intelligence if performed by humans—is the most widely recognized. Defining intelligence is not simple; generally speaking, intelligence is a collection of skills including comprehension, learning, and problem-solving thinking. AI uses a variety of technologies to mimic these facets of human intellect. [ 2 ] In its widest definition, AI refers to the intelligence displayed by machines, especially computer systems. Since its inception as an area of study in 1956, AI has seen several cycles of optimism, followed by depressing and financially debilitating times known as AI winters. After deep learning surpassed earlier AI methods in 2012, funding and attention skyrocketed. John McCarthy, Marvin Minksy, Nathaniel Rochester, and Claude Shannon—now regarded as the founding fathers of artificial intelligence. [ 3 ] Their work laid the foundation for the development of AI technologies that continue to advance today. Science and technology are always evolving, and the application of artificial intelligence in health care is becoming more widely recognized.[ 4 ] Developed countries have put a lot of money into AI and healthcare applications. [ 1 ] Since the introduction of COVID-19, the need for artificial intelligence expertise and resources in the health industry has skyrocketed in order to minimize workload and diagnostic errors.[ 5 ] The implementation of artificial intelligence (AI) in tertiary care institutions is critical for altering health care and patient outcomes. AI may use its capacity to handle massive amounts of data, automate workflows, and give rapid results. As a result, artificial intelligence will help improve clinical decision-making, hospital management, and patient care.[ 6 ] AI in healthcare is not just about new technology; it's also about addressing the growing problems in the field. [ 7 ] AI can analyse test data, scans, and medical records to identify trends and issues that a doctor would have overlooked, resulting in a quicker and more precise diagnosis. [ 8 ] Not only can AI help with diagnosis, but it can also evaluate a patient's vital signs and monitor any abnormalities that may indicate a health problem. This enables the physician to respond swiftly and avoid major potential health complications, hence enhancing overall patient safety. [ 9 ] The advantages of AI in healthcare are being observed in developed nations. [ 10 ] According to a research, by 2026, AI may save the US healthcare system over $ 150 billion annually. [ 11 ] Consequently, Canadian and American medical schools have developed new curricular efforts to meet the skills necessary for success in medicine in the twenty-first century. Additionally, the University Health Network of Canada constructed a machine learning system to build radiation therapy plans based on patient data. [ 12 ] In nations such as Iraq and Saudi Arabia, artificial intelligence is employed for not only image processing, but also for prognosis, therapy, and patient monitoring. [ 13 ] In countries with healthcare systems that struggle with unequal resource distribution, shortages of staff, and an increase in the burden of sickness, artificial intelligence is typically seen as the answer to these problems.[ 14 ] The gap between expectations and implementation, however, reveals major obstacles such inadequate infrastructure, moral dilemmas, and the requirement for regionally tailored solutions. [ 10 ] AI can predict outbreaks like malaria, [ 15 ] diagnose eye diseases through smartphone-based screening, [ 16 ] and help automate tasks like managing patient records, which lowers costs. [ 10 ] The system has multiple health issues that include, but are not limited to the geographic dispersion of health facilities and the inadequate number of skilled health personnel. An implementation of AI in healthcare processes should enable such issues to be resolved through the augmentation of accessibility, precision and efficiency of health services. However, Pakistan is still in the early phases of implementing health informatics.[ 17 ] AI development has been growing in the country since 2018, as part of the president's AI and computing strategy, which aims to foster innovation and technology in a variety of industries, including healthcare.[ 1 ] Research conducted in Pakistan has demonstrated that AI is improving in the healthcare sector. According to an online study conducted by Zaboor Ahmed et al., the majority of medical students and clinicians have a positive attitude about the usage of AI, saying it may help with diagnosis, therapy, and patient monitoring. [ 1 ] Another study indicated that AI training programs can increase healthcare students' knowledge, abilities, and attitudes about AI. [ 18 ] Similarly, a review of telemedicine found that AI may assist doctors and patients in remote places, making it simpler for individuals to access healthcare. [ 18 ] According to a research, 74% of Pakistani doctors are familiar with AI, 27.3% are aware of its usage in medicine, and 76.7% have a good view about its use.[ 19 ] According to a study In KPK the 91.2% of the doctors of KPK have knowledge about AI but only 30.6% use it in their practice. [ 20 ] This study aims to evaluate doctors' perceptions of AI in healthcare and their knowledge of AI in the medical field at Peshawar's tertiary care hospitals. It is assumed that the vast majority of Peshawar's doctors do not utilize AI in their patient treatment. Material and methods A descriptive, cross-sectional design was used to carry out the study. The target population consisted of physicians employed at Peshawar's tertiary care facilities. The following formula was used to get the sample size: $$\:n=\frac{{Z}^{2}.p.\left(1-p\right)}{{E}^{2}}$$ The sample size was determined using a 95% confidence level and a 5% margin of error. Physicians were chosen using simple selection from a population of 4758 in the Peshawar area. The research comprised physicians from Peshawar's tertiary care institutions. MBBS degrees were utilized to identify participation, while paramedic workers were excluded. A questionnaire was utilized to collect data prospectively across five months, from April to August 2024. All data was collected at some point in time. To collect data, the selected sample was given a closed-ended questionnaire as well as a structured questionnaire to complete. The questionnaire was designed to collect information on the doctors' demographics, titles, level of AI awareness, and perspectives on the research subject. After data collection, any inconsistencies or missing replies were discarded. The full data set was coded and analysed using SPSS Version 19. Descriptive statistics were used to provide a description of the sample's key features, such as age and specialization. The population's awareness was measured using univariate descriptive statistics, and the relationship between the variables of interest was investigated using exploratory factor analysis. With p-values less than 0.001 indicating a significant link, the data were interpreted in terms of statistical significance. The study complied with all ethical standards for research involving human subjects. Throughout the poll, each participant received informed permission and a guarantee of their privacy. The data collected was kept private and used exclusively for this research. The study was carried out on its own without any other monetary support. No firm, institute, or organization has contributed funds or assistance. No conflicts of interest are disclosed by the writers. The writers' ideas are their own, and no financial or personal ties have influenced them. Results The demographic data indicated that most respondents were men (70.5%) and women 29.5%. The remaining samples by profession were House Officers (34.5%), Medical Officers (24.1%), and Trainee Medical Officers (23.8%). Postgraduate Residents were 11.6% and the least number of respondents were Consultants (4.4%) and Professors (1.6%). Overall, it appears that the sample is largely made up by junior medical practitioners than this that were more senior. As shown in Table 1 Table 1 Demographic Profile of the Respondents Demographic Questions Category Frequency Percentage Cumulative Percentage Gender Male 255 70.50% 70.50% Female 94 29.50% 100.00% Professional Role House Officers 110 34.50% 34.50% Medical Officers 77 24.10% 58.60% Trainee Medical Officers 76 23.80% 82.40% Postgraduate Residents 37 11.60% 94.00% Consultants 14 4.40% 98.40% Professors 5 1.60% 100.00% The univariate descriptive statistics show that more than nine out of ten respondents (91.9%) said they understood AI, while 8.1% didn't; the mean awareness score was 1.08 (SD = 0.27), with both the mean and median at 1.00, indicating that respondents largely agreed with knowing. Unique to note is that whereas 40.6% were found to use AI in the care of their patients, however 59.4% were not. The mean AI usage in patient care was 1.59 (SD = 0.49), with both the median and mode at 2.00, indicated that most of the respondents in this study were resistant to utilize AI in clinical practice regardless of the amount of knowledge they possessed about the technology. As illustrated in 2 Table 2 Univariate Descriptive Analysis Variables Category Frequency Percentage (%) Statistical Measures Awareness of AI Aware 294 91.90% Mean = 1.08 (SD = 0.27) Not Aware 26 8.10% Median = 1.00, Mode = 1.00 Use of AI in Patient Care Uses AI for Patient Care 130 40.60% Mean = 1.59 (SD = 0.49) Does Not Use AI 190 59.40% Median = 2.00, Mode = 2.00 The exploratory factor analysis shows moderate sampling adequacy, with a KMO of 0.500 and a significant Bartlett's Test of Sphericity (χ² = 7.504, p = 0.006), indicating that the data is suitable for factor analysis. One main factor was found, accounting for 57.642% of total variance, along with high factor loadings (0.75) between "Use of AI for patient care" and "Information about AI", indicating that the two variables are related. The pattern matrix also indicated one primary component - the results do not suggest further splitting of the factors. The correlation study found a small but statistically significant positive correlation (r = 0.15, p = 0.013), indicating that higher understanding of AI is associated with a slight increase in its use in patient care. Further analyses, such as Fisher's Exact Test (p-values = 0.02 and < 0.001), attest to the strong correlation between AI knowledge and its implementation. This suggests that doctors who know about AI are more likely to incorporate it into clinic practice, even though other determinants may be at play. As illustrated in Table 3 Table 3 Exploratory Factor Analysis Test Type of Test Values Sampling Adequacy Test Kaiser-Meyer-Olkin (KMO) Measure 0.50 (Moderate sampling adequacy) Bartlett's Test of Sphericity χ² = 7.504, p = 0.006 (Significant relationship) Extracted Communities 0.57 ("Use of AI for patient care" & "Information about AI") Total Variance Explained Total Variance Explained 57.642% (Majority of variation assigned to one factor) Factor Cross Loadings Factor Loadings 0.75 (Strong connection between AI information & AI use) Pattern Matrix Identified Factors 1 (No further factor adjudication needed) Correlation Analysis Correlation Coefficient (r) 0.15 (Weak but significant positive correlation) Significance (p-value) 0.013 (Statistically significant) Interpretations Increased AI information slightly increases AI use in patient care, but other factors likely contribute Summary Statistics Probability Test 6.96, p = 0.008 (Significant association) Fisher’s Exact Test (2-sided) p = 0.02 Fisher’s Exact Test (1-sided) p < 0.001 Interpretation Doctors knowledgeable about AI are more likely to use it in patient care The descriptive statistics highlight that the most commonly reported difficulty in adopting AI in clinical settings was a lack of training and knowledge, cited by 33.9% of respondents Next came several, combined issues (24.1%), indicating that many persons experienced related challenges rather than a difficulty with a single item. Infrastructure difficulties, such as cost and access, accounted for 17.6%, while technical issues were reported by 13.8%. A lesser proportion of respondents (7.5%) recognized insufficient medical knowledge as a barrier, whereas 3.1% reported having no problems at all. Overall, it appears that the most significant impediments to managing patient care using AI were inadequate training and multidimensional problems. As illustrated in Table 4 Table 4 Descriptive Statistics Difficulty Frequency Percentage Lack of training and knowledge 108 33.90% Lack of infrastructure (limited access/cost) 56 17.60% Technical issues 44 13.80% Poor medical knowledge 24 7.50% No difficulty reported 10 3.10% Multiple challenges combined 77 24.10% Total 319 100% Discussion Numerous research studies on the use of AI in healthcare have been conducted; however, the majorities of them were based only on radiology and were conducted on medical students and paramedic staff. This makes it necessary for us to conduct this study on the usage of AI by physicians at Peshawar's tertiary care institutions. The study highlights Peshawar's tertiary care facilities' understanding, utilization, and barriers to applying artificial intelligence into clinical operations. The findings, however, indicated a significant contradiction: while a considerable proportion of physicians (91.9%) were aware of AI, just 40.6% employed it for patient treatment. This conclusion is consistent with another poll of Khyber Pakhtunkhwa doctors, which found that 91.2% are familiar with artificial intelligence.[ 21 ] This disparity between high understanding and low usage demonstrates that technology does not necessarily transfer into application, especially in situations with limited resources. Even while many AI applications have already been deployed in affluent nations, this contrasts sharply with situations with limited resources, where AI applications are still in their early stages. [ 10 ]. Our study on patient care revealed that, whereas many clinicians were aware of AI, only 40.6% of them reported using it to treat patients.[ 21 ] Similarly, a survey conducted in Peshawar found that just 11.3% of respondents used AI for practical applications, while admitting its potential to simplify clinical duties.[ 1 ] Our study was conducted with a large number of early career professionals such as house officers and trainee medical officers, indicating that despite being increasingly technologically exposed, practical integration remains limited. Even though education is the first step, it is insufficient to drive integration, as evidenced by the significant but weakly positive relationship found by exploratory factor analysis and correlation analysis between awareness of AI and its use in clinical practice (r = 0.151, p = 0.013). This is consistent with research that highlights the importance of perceived usefulness in technology, workflow compatibility, and institutional support.[ 22 , 23 ] This discrepancy can be further explained by the high percentage of early career professionals—34.5% of the house office in our study—who may not have the independence to deploy AI technologies without senior support or infrastructure. Lack of experience and training was the most often mentioned obstacle to adopting AI for medical care, accounting for 33.9% of responses. These findings are consistent with global research, which indicates that inadequate training remains a concern, particularly in resource-rich places.[ 23 ] These findings are consistent with global research, which indicates that inadequate training remains a concern, particularly in resource-rich places.[31] It's worth noting that technical concerns (13.8%) and a lack of medical experience (7.5%) were mentioned less frequently, indicating that physicians value developing their skills above instrument reliability. These problems, when considered together, underscore the need for a range of interventions that target both institutional preparation and individual skills. It was fascinating to see that just 3.1% of respondents had no difficulty using AI, underscoring the relevance of these concerns. Moderate sampling adequacy (KMO = 0.500) supports the need for bigger studies with more diverse samples to give deeper insights into the challenges faced by various doctors. The findings suggest that AI education and infrastructure development should be a primary focus for policymakers and healthcare institutions. It should be possible for many experts to participate in training sessions and seminars in order to bridge the knowledge and practice gap. Addressing structural concerns such as high pricing, a lack of technical assistance, and uneven access to AI technology can also help increase adoption rates. The study's cross-sectional design limits its ability to determine causality; the study was limited to Peshawar's tertiary care hospitals, so the findings are not generalizable; the data was self-reported, the study was limited to Peshawar's tertiary care hospitals, so the findings are not generalizable; the data was self-reported, limiting detailed insights; and some barriers to AI adoption may have gone unnoticed. Future study employing longitudinal designs, qualitative approaches, and a broader geographic scope will give a more in-depth knowledge of these components. Conclusion The study shows that there is a gap between the use of AI in patient care and its awareness among the doctors of tertiary care hospitals in Peshawar. Even though a lot of doctors recognize the importance of AI only a few use it in clinical practice for patient care, this may be because of barriers like lack of infrastructure and training. The link shows that knowledge alone isn’t enough, institutional support and practical training are important too. Thus it is important for medical institutions to invest in training and technology to improve clinical practice. Future research should examine the effect of various barriers on the use of AI in patient care, and training outcomes, and explore the role of AI in multiple specialties. Declarations Acknowledgement: We express a deep gratitude to Sir Dr. Rizwan ul Haq for their guidance and continuous support. We are also thankful to the research committee of Jinnah Medical College for their approval, constructive feedback, and their support in facilitating this study. Disclaimer: The authors declare no disclaimer regarding this study. Conflict of interest: The authors disclose no conflicts of interest. All opinions are those of the writers and are not affected by any personal or financial affiliations. Funding disclosure: The study was done without any funding assistance from an outside source. No money or support has been provided by any organization, institute, or corporation. Author Contribution A.F., A.R. and H.U. wrote the main manuscrript, A.F. and A.R. also did the data compilation and Analysis , while Q.Z. and A.Q collected the required data for the manuscript while R.H. supervised the whole research. References Ahmed, Z., et al., Knowledge, attitude, and practice of artificial intelligence among doctors and medical students in Pakistan: A cross-sectional online survey . Ann Med Surg (Lond), 2022. 76: p. 103493. Rigla, M., et al., Artificial Intelligence Methodologies and Their Application to Diabetes . J Diabetes Sci Technol, 2018. 12(2): p. 303–310. Sadiq, F., et al., Knowledge, Attitude, and Practice (KAP) Regarding the Use of Artificial Intelligence in Hospital Settings in Mardan, Khyber Pakhtunkhwa, Pakistan . Cureus, 2024. 16(12): p. e75355. Xie, Y., Y. Zhai, and G. Lu, Evolution of artificial intelligence in healthcare: a 30-year bibliometric study . Frontiers in Medicine, 2025. Volume 11–2024. Alwadani, F.A.S., et al., Attitude and Understanding of Artificial Intelligence Among Saudi Medical Students: An Online Cross-Sectional Study . J Multidiscip Healthc, 2024. 17: p. 1887–1899. Bhagat, S.V. and D. Kanyal, Navigating the Future: The Transformative Impact of Artificial Intelligence on Hospital Management- A Comprehensive Review . Cureus, 2024. 16(2): p. e54518. Chen, M. and M. Decary, Artificial intelligence in healthcare: An essential guide for health leaders . Healthc Manage Forum, 2020. 33(1): p. 10–18. Maleki Varnosfaderani, S. and M. Forouzanfar, The Role of AI in Hospitals and Clinics: Transforming Healthcare in the 21st Century . Bioengineering (Basel), 2024. 11(4). Giordano, C., et al., Accessing Artificial Intelligence for Clinical Decision-Making . Frontiers in Digital Health, 2021. Volume 3–2021. Brian, W., et al., Artificial intelligence (AI) and global health: how can AI contribute to health in resource-poor settings? BMJ Global Health, 2018. 3(4): p. e000798. Fei, J., et al., Artificial intelligence in healthcare: past, present and future . Stroke and Vascular Neurology, 2017. 2(4). Li, D., K. Kulasegaram, and B.D. Hodges, Why We Needn't Fear the Machines: Opportunities for Medicine in a Machine Learning World . Acad Med, 2019. 94(5): p. 623–625. Boillat, T., F.A. Nawaz, and H. Rivas, Readiness to Embrace Artificial Intelligence Among Medical Doctors and Students: Questionnaire-Based Study . JMIR Med Educ, 2022. 8(2): p. e34973. Boutayeb, A., The double burden of communicable and non-communicable diseases in developing countries . Trans R Soc Trop Med Hyg, 2006. 100(3): p. 191–9. Lasrado, N., et al., Attenuated strain of CVB3 with a mutation in the CAR-interacting region protects against both myocarditis and pancreatitis . Sci Rep, 2021. 11(1): p. 12432. Tufail, A., et al., Automated Diabetic Retinopathy Image Assessment Software: Diagnostic Accuracy and Cost-Effectiveness Compared with Human Graders . Ophthalmology, 2017. 124(3): p. 343–351. Patoli, A.Q., Role of syndromic management using dynamic machine learning in future of e-Health in Pakistan . Stud Health Technol Inform, 2007. 129(Pt 1): p. 601–4. Malik, M., A.F. Kazi, and A. Hussain, Adoption of health technologies for effective health information system: Need of the hour for Pakistan . PLoS One, 2021. 16(10): p. e0258081. Sajjad, W., et al., Knowledge, attitude, and practices regarding use of artificial intelligence for medical writings among doctors of Khyber Pakhtunkhwa, Pakistan: a cross-sectional study . Ann Med Surg (Lond), 2025. 87(3): p. 1190–1199. Matuchansky, C., Deep medicine, artificial intelligence, and the practising clinician . Lancet, 2019. 394(10200): p. 736. Sajjad, W., et al., Knowledge, attitude, and practices regarding use of artificial intelligence for medical writings among doctors of Khyber Pakhtunkhwa, Pakistan: a cross-sectional study . Annals of Medicine and Surgery, 2025. 87(3): p. 1190–1199. Jeilani, A. and S. Abubakar, Perceived institutional support and its effects on student perceptions of AI learning in higher education: the role of mediating perceived learning outcomes and moderating technology self-efficacy . Frontiers in Education, 2025. Volume 10–2025. Al-Abdullatif, A.M., Modeling Teachers’ Acceptance of Generative Artificial Intelligence Use in Higher Education: The Role of AI Literacy, Intelligent TPACK, and Perceived Trust . Education Sciences, 2024. 14(11): p. 1209. 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-7290827","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":499624358,"identity":"9d130484-fe94-4be4-8226-1c6c2557d01f","order_by":0,"name":"Amna Faiz","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA/ElEQVRIiWNgGAWjYLCCCgYJBn5mhsQHH4AcNnZitJwBapFsb3hsOAOkhZk4LQwMBmcOPhPmAfEIaeGfdvbghwM1FvYGN5LTmG1+bZPnY2Zg/PAxB7cWidt5yRIHjkkwS95IS3uc23fbsI2ZgVly5jY81tzOMZD+wCbBxncjJ904t+c2I1ALGzMvHi3yt3OMfxz4J8HDcCP/m7Rlz217gloMbueYSRxsk5AQOHMgTZrhx+1EgloMgVosDvZJGAADOdmwt+F2chszYzNev8gBHXbjwLc6e3BU/vhz23Z+e/PBDx/xeR8FMLaByQZi1YPAH1IUj4JRMApGwUgBAJ3SVCz15E9FAAAAAElFTkSuQmCC","orcid":"","institution":"","correspondingAuthor":true,"prefix":"","firstName":"Amna","middleName":"","lastName":"Faiz","suffix":""},{"id":499624359,"identity":"564e9368-aacc-45d3-a60d-168b7c5c7475","order_by":1,"name":"Rizwan Ul Haq","email":"","orcid":"","institution":"","correspondingAuthor":false,"prefix":"","firstName":"Rizwan","middleName":"Ul","lastName":"Haq","suffix":""},{"id":499624360,"identity":"351d23ef-b7df-43ae-87d4-4dedb0c56c19","order_by":2,"name":"Anees Ur Rehman","email":"","orcid":"","institution":"","correspondingAuthor":false,"prefix":"","firstName":"Anees","middleName":"Ur","lastName":"Rehman","suffix":""},{"id":499624361,"identity":"cc1f10ba-eee3-4129-9224-2eeadf75b95b","order_by":3,"name":"Hammad Ullah","email":"","orcid":"","institution":"","correspondingAuthor":false,"prefix":"","firstName":"Hammad","middleName":"","lastName":"Ullah","suffix":""},{"id":499624362,"identity":"01d53dc2-5e7e-4490-94e7-d1cf7080d94a","order_by":4,"name":"Quratulain Zahir","email":"","orcid":"","institution":"","correspondingAuthor":false,"prefix":"","firstName":"Quratulain","middleName":"","lastName":"Zahir","suffix":""},{"id":499624363,"identity":"e772d675-4a3d-4652-a1ce-f681a051a918","order_by":5,"name":"Ahsan Qayyum","email":"","orcid":"","institution":"","correspondingAuthor":false,"prefix":"","firstName":"Ahsan","middleName":"","lastName":"Qayyum","suffix":""}],"badges":[],"createdAt":"2025-08-04 11:53:19","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-7290827/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-7290827/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":89972269,"identity":"78ae6f3b-73d5-467b-a10d-6a161d6c7bdd","added_by":"auto","created_at":"2025-08-27 05:39:43","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":507710,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7290827/v1/cc2555b5-88d0-4c69-b8f0-56429ff966b7.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"Utilization of AI by the doctors for patient care in tertiary care hospital Peshawar","fulltext":[{"header":"Introduction","content":"\u003cp\u003eArtificial intelligence (AI) has drawn extraordinary interest in recent decades and is being referred to be the fourth industrial revolution.[\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e] There are several definitions of AI, however, at the moment, Boden's definition\u0026mdash;the capacity to enable computers to perform tasks that would need intelligence if performed by humans\u0026mdash;is the most widely recognized. Defining intelligence is not simple; generally speaking, intelligence is a collection of skills including comprehension, learning, and problem-solving thinking. AI uses a variety of technologies to mimic these facets of human intellect. [\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e] In its widest definition, AI refers to the intelligence displayed by machines, especially computer systems.\u003c/p\u003e\u003cp\u003eSince its inception as an area of study in 1956, AI has seen several cycles of optimism, followed by depressing and financially debilitating times known as AI winters. After deep learning surpassed earlier AI methods in 2012, funding and attention skyrocketed. John McCarthy, Marvin Minksy, Nathaniel Rochester, and Claude Shannon\u0026mdash;now regarded as the founding fathers of artificial intelligence. [\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e] Their work laid the foundation for the development of AI technologies that continue to advance today.\u003c/p\u003e\u003cp\u003eScience and technology are always evolving, and the application of artificial intelligence in health care is becoming more widely recognized.[\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e] Developed countries have put a lot of money into AI and healthcare applications. [\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e] Since the introduction of COVID-19, the need for artificial intelligence expertise and resources in the health industry has skyrocketed in order to minimize workload and diagnostic errors.[\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e]\u003c/p\u003e\u003cp\u003eThe implementation of artificial intelligence (AI) in tertiary care institutions is critical for altering health care and patient outcomes. AI may use its capacity to handle massive amounts of data, automate workflows, and give rapid results. As a result, artificial intelligence will help improve clinical decision-making, hospital management, and patient care.[\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e] AI in healthcare is not just about new technology; it's also about addressing the growing problems in the field. [\u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e] AI can analyse test data, scans, and medical records to identify trends and issues that a doctor would have overlooked, resulting in a quicker and more precise diagnosis. [\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e] Not only can AI help with diagnosis, but it can also evaluate a patient's vital signs and monitor any abnormalities that may indicate a health problem. This enables the physician to respond swiftly and avoid major potential health complications, hence enhancing overall patient safety. [\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e]\u003c/p\u003e\u003cp\u003eThe advantages of AI in healthcare are being observed in developed nations. [\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e] According to a research, by 2026, AI may save the US healthcare system over \u003cspan\u003e$\u003c/span\u003e150\u0026nbsp;billion annually. [\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e] Consequently, Canadian and American medical schools have developed new curricular efforts to meet the skills necessary for success in medicine in the twenty-first century. Additionally, the University Health Network of Canada constructed a machine learning system to build radiation therapy plans based on patient data. [\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e]\u003c/p\u003e\u003cp\u003eIn nations such as Iraq and Saudi Arabia, artificial intelligence is employed for not only image processing, but also for prognosis, therapy, and patient monitoring. [\u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e] In countries with healthcare systems that struggle with unequal resource distribution, shortages of staff, and an increase in the burden of sickness, artificial intelligence is typically seen as the answer to these problems.[\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e] The gap between expectations and implementation, however, reveals major obstacles such inadequate infrastructure, moral dilemmas, and the requirement for regionally tailored solutions. [\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e] AI can predict outbreaks like malaria, [\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e] diagnose eye diseases through smartphone-based screening, [\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e] and help automate tasks like managing patient records, which lowers costs. [\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e]\u003c/p\u003e\u003cp\u003eThe system has multiple health issues that include, but are not limited to the geographic dispersion of health facilities and the inadequate number of skilled health personnel. An implementation of AI in healthcare processes should enable such issues to be resolved through the augmentation of accessibility, precision and efficiency of health services. However, Pakistan is still in the early phases of implementing health informatics.[\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e] AI development has been growing in the country since 2018, as part of the president's AI and computing strategy, which aims to foster innovation and technology in a variety of industries, including healthcare.[\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e]\u003c/p\u003e\u003cp\u003eResearch conducted in Pakistan has demonstrated that AI is improving in the healthcare sector. According to an online study conducted by Zaboor Ahmed et al., the majority of medical students and clinicians have a positive attitude about the usage of AI, saying it may help with diagnosis, therapy, and patient monitoring. [\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e] Another study indicated that AI training programs can increase healthcare students' knowledge, abilities, and attitudes about AI. [\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e] Similarly, a review of telemedicine found that AI may assist doctors and patients in remote places, making it simpler for individuals to access healthcare. [\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e]\u003c/p\u003e\u003cp\u003e According to a research, 74% of Pakistani doctors are familiar with AI, 27.3% are aware of its usage in medicine, and 76.7% have a good view about its use.[\u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e] According to a study In KPK the 91.2% of the doctors of KPK have knowledge about AI but only 30.6% use it in their practice. [\u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e]\u003c/p\u003e\u003cp\u003eThis study aims to evaluate doctors' perceptions of AI in healthcare and their knowledge of AI in the medical field at Peshawar's tertiary care hospitals. It is assumed that the vast majority of Peshawar's doctors do not utilize AI in their patient treatment.\u003c/p\u003e"},{"header":"Material and methods","content":"\u003cp\u003eA descriptive, cross-sectional design was used to carry out the study. The target population consisted of physicians employed at Peshawar's tertiary care facilities. The following formula was used to get the sample size:\u003cdiv id=\"Equa\" class=\"Equation\"\u003e\u003cdiv format=\"TEX\" class=\"mathdisplay\" id=\"FileID_Equa\" name=\"EquationSource\"\u003e\n$$\\:n=\\frac{{Z}^{2}.p.\\left(1-p\\right)}{{E}^{2}}$$\u003c/div\u003e\u003c/div\u003e\u003c/p\u003e\u003cp\u003eThe sample size was determined using a 95% confidence level and a 5% margin of error. Physicians were chosen using simple selection from a population of 4758 in the Peshawar area. The research comprised physicians from Peshawar's tertiary care institutions. MBBS degrees were utilized to identify participation, while paramedic workers were excluded.\u003c/p\u003e\u003cp\u003eA questionnaire was utilized to collect data prospectively across five months, from April to August 2024. All data was collected at some point in time. To collect data, the selected sample was given a closed-ended questionnaire as well as a structured questionnaire to complete. The questionnaire was designed to collect information on the doctors' demographics, titles, level of AI awareness, and perspectives on the research subject.\u003c/p\u003e\u003cp\u003eAfter data collection, any inconsistencies or missing replies were discarded. The full data set was coded and analysed using SPSS Version 19. Descriptive statistics were used to provide a description of the sample's key features, such as age and specialization.\u003c/p\u003e\u003cp\u003eThe population's awareness was measured using univariate descriptive statistics, and the relationship between the variables of interest was investigated using exploratory factor analysis. With p-values less than 0.001 indicating a significant link, the data were interpreted in terms of statistical significance.\u003c/p\u003e\u003cp\u003e The study complied with all ethical standards for research involving human subjects. Throughout the poll, each participant received informed permission and a guarantee of their privacy. The data collected was kept private and used exclusively for this research.\u003c/p\u003e\u003cp\u003eThe study was carried out on its own without any other monetary support. No firm, institute, or organization has contributed funds or assistance. No conflicts of interest are disclosed by the writers. The writers' ideas are their own, and no financial or personal ties have influenced them.\u003c/p\u003e"},{"header":"Results","content":"\u003cp\u003eThe demographic data indicated that most respondents were men (70.5%) and women 29.5%. The remaining samples by profession were House Officers (34.5%), Medical Officers (24.1%), and Trainee Medical Officers (23.8%). Postgraduate Residents were 11.6% and the least number of respondents were Consultants (4.4%) and Professors (1.6%). Overall, it appears that the sample is largely made up by junior medical practitioners than this that were more senior. As shown in Table \u003cspan class=\"InternalRef\"\u003e1\u003c/span\u003e\u003c/p\u003e\n\u003cdiv class=\"gridtable\"\u003e\n \u003ctable id=\"Tab1\" border=\"1\"\u003e\n \u003ccaption\u003e\n \u003cdiv class=\"CaptionNumber\"\u003eTable 1\u003c/div\u003e\n \u003cdiv class=\"CaptionContent\"\u003e\n \u003cp\u003eDemographic Profile of the Respondents\u003c/p\u003e\n \u003c/div\u003e\n \u003c/caption\u003e\n \u003cthead\u003e\n \u003ctr\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003eDemographic Questions\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003eCategory\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003eFrequency\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003ePercentage\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003eCumulative Percentage\u003c/p\u003e\n \u003c/th\u003e\n \u003c/tr\u003e\n \u003c/thead\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd rowspan=\"2\" align=\"left\"\u003e\n \u003cp\u003eGender\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eMale\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e255\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e70.50%\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e70.50%\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eFemale\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e94\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e29.50%\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e100.00%\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd rowspan=\"6\" align=\"left\"\u003e\n \u003cp\u003eProfessional Role\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eHouse Officers\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e110\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e34.50%\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e34.50%\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eMedical Officers\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e77\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e24.10%\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e58.60%\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eTrainee Medical Officers\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e76\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e23.80%\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e82.40%\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003ePostgraduate Residents\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e37\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e11.60%\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e94.00%\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eConsultants\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e14\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e4.40%\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e98.40%\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eProfessors\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e5\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e1.60%\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e100.00%\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n \u003c/table\u003e\n \u003cp\u003eThe univariate descriptive statistics show that more than nine out of ten respondents (91.9%) said they understood AI, while 8.1% didn\u0026apos;t; the mean awareness score was 1.08 (SD = 0.27), with both the mean and median at 1.00, indicating that respondents largely agreed with knowing. Unique to note is that whereas 40.6% were found to use AI in the care of their patients, however 59.4% were not. The mean AI usage in patient care was 1.59 (SD = 0.49), with both the median and mode at 2.00, indicated that most of the respondents in this study were resistant to utilize AI in clinical practice regardless of the amount of knowledge they possessed about the technology. \u0026nbsp;As illustrated in 2\u003c/p\u003e\n\u003c/div\u003e\n\u003cdiv class=\"gridtable\"\u003e\n \u003cdiv class=\"colspec\" align=\"left\"\u003e\u0026nbsp;\u003c/div\u003e\n \u003cdiv class=\"colspec\" align=\"left\"\u003e\u0026nbsp;\u003c/div\u003e\n\u003c/div\u003e\n\u003cdiv class=\"gridtable\"\u003e\n \u003ctable id=\"Tab3\" border=\"1\"\u003e\n \u003ccaption\u003e\n \u003cdiv class=\"CaptionNumber\"\u003eTable 2\u003c/div\u003e\n \u003cdiv class=\"CaptionContent\"\u003e\n \u003cp\u003eUnivariate Descriptive Analysis\u003c/p\u003e\n \u003c/div\u003e\n \u003c/caption\u003e\n \u003cthead\u003e\n \u003ctr\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003eVariables\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003eCategory\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003eFrequency\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003ePercentage (%)\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003eStatistical Measures\u003c/p\u003e\n \u003c/th\u003e\n \u003c/tr\u003e\n \u003c/thead\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd rowspan=\"2\" align=\"left\"\u003e\n \u003cp\u003eAwareness of AI\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eAware\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e294\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e91.90%\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eMean\u0026thinsp;=\u0026thinsp;1.08 (SD\u0026thinsp;=\u0026thinsp;0.27)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eNot Aware\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e26\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e8.10%\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eMedian\u0026thinsp;=\u0026thinsp;1.00, Mode\u0026thinsp;=\u0026thinsp;1.00\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd rowspan=\"2\" align=\"left\"\u003e\n \u003cp\u003eUse of AI in Patient Care\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eUses AI for Patient Care\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e130\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e40.60%\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eMean\u0026thinsp;=\u0026thinsp;1.59 (SD\u0026thinsp;=\u0026thinsp;0.49)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eDoes Not Use AI\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e190\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e59.40%\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eMedian\u0026thinsp;=\u0026thinsp;2.00, Mode\u0026thinsp;=\u0026thinsp;2.00\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n \u003c/table\u003e\n\u003c/div\u003e\n\u003cp\u003eThe exploratory factor analysis shows moderate sampling adequacy, with a KMO of 0.500 and a significant Bartlett\u0026apos;s Test of Sphericity (\u0026chi;\u0026sup2; = 7.504, p\u0026thinsp;=\u0026thinsp;0.006), indicating that the data is suitable for factor analysis.\u003c/p\u003e\n\u003cp\u003eOne main factor was found, accounting for 57.642% of total variance, along with high factor loadings (0.75) between \u0026quot;Use of AI for patient care\u0026quot; and \u0026quot;Information about AI\u0026quot;, indicating that the two variables are related. The pattern matrix also indicated one primary component - the results do not suggest further splitting of the factors. The correlation study found a small but statistically significant positive correlation (r\u0026thinsp;=\u0026thinsp;0.15, p\u0026thinsp;=\u0026thinsp;0.013), indicating that higher understanding of AI is associated with a slight increase in its use in patient care.\u003c/p\u003e\n\u003cp\u003eFurther analyses, such as Fisher\u0026apos;s Exact Test (p-values\u0026thinsp;=\u0026thinsp;0.02 and \u0026lt;\u0026thinsp;0.001), attest to the strong correlation between AI knowledge and its implementation. This suggests that doctors who know about AI are more likely to incorporate it into clinic practice, even though other determinants may be at play. As illustrated in Table\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e3\u003c/span\u003e\u003c/p\u003e\n\u003cdiv class=\"gridtable\"\u003e\n \u003cdiv class=\"colspec\" align=\"left\"\u003e\u0026nbsp;\u003c/div\u003e\n \u003ctable id=\"Tab4\" border=\"1\"\u003e\n \u003ccaption\u003e\n \u003cdiv class=\"CaptionNumber\"\u003eTable 3\u003c/div\u003e\n \u003cdiv class=\"CaptionContent\"\u003e\n \u003cp\u003e\u003cem\u003eExploratory Factor Analysis\u003c/em\u003e\u003c/p\u003e\n \u003c/div\u003e\n \u003c/caption\u003e\n \u003cthead\u003e\n \u003ctr\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003eTest\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003eType of Test\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003eValues\u003c/p\u003e\n \u003c/th\u003e\n \u003c/tr\u003e\n \u003c/thead\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd rowspan=\"2\" align=\"left\"\u003e\n \u003cp\u003eSampling Adequacy Test\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eKaiser-Meyer-Olkin (KMO) Measure\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.50 (Moderate sampling adequacy)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eBartlett\u0026apos;s Test of Sphericity\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u0026chi;\u0026sup2; = 7.504, p\u0026thinsp;=\u0026thinsp;0.006 (Significant relationship)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eExtracted Communities\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.57 (\u0026quot;Use of AI for patient care\u0026quot; \u0026amp; \u0026quot;Information about AI\u0026quot;)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eTotal Variance Explained\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eTotal Variance Explained\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e57.642% (Majority of variation assigned to one factor)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eFactor Cross Loadings\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eFactor Loadings\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.75 (Strong connection between AI information \u0026amp; AI use)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003ePattern Matrix\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eIdentified Factors\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e1 (No further factor adjudication needed)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd rowspan=\"2\" align=\"left\"\u003e\n \u003cp\u003eCorrelation Analysis\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eCorrelation Coefficient (r)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.15 (Weak but significant positive correlation)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eSignificance (p-value)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.013 (Statistically significant)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eInterpretations\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eIncreased AI information slightly increases AI use in patient care, but other factors likely contribute\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd rowspan=\"4\" align=\"left\"\u003e\n \u003cp\u003eSummary Statistics\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eProbability Test\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e6.96, p\u0026thinsp;=\u0026thinsp;0.008 (Significant association)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eFisher\u0026rsquo;s Exact Test (2-sided)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003ep\u0026thinsp;=\u0026thinsp;0.02\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eFisher\u0026rsquo;s Exact Test (1-sided)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003ep\u0026thinsp;\u0026lt;\u0026thinsp;0.001\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eInterpretation\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eDoctors knowledgeable about AI are more likely to use it in patient care\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n \u003c/table\u003e\n\u003c/div\u003e\n\u003cdiv class=\"gridtable\"\u003e\n \u003cdiv class=\"colspec\" align=\"left\"\u003e\u0026nbsp;\u003c/div\u003e\n \u003cdiv class=\"colspec\" align=\"left\"\u003e\u0026nbsp;\u003c/div\u003e\n \u003cdiv class=\"colspec\" align=\"left\"\u003eThe descriptive statistics highlight that the most commonly reported difficulty in adopting AI in clinical settings was a lack of training and knowledge, cited by 33.9% of respondents Next came several, combined issues (24.1%), indicating that many persons experienced related challenges rather than a difficulty with a single item. Infrastructure difficulties, such as cost and access, accounted for 17.6%, while technical issues were reported by 13.8%. A lesser proportion of respondents (7.5%) recognized insufficient medical knowledge as a barrier, whereas 3.1% reported having no problems at all. Overall, it appears that the most significant impediments to managing patient care using AI were inadequate training and multidimensional problems. As illustrated in Table\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e4\u003c/span\u003e\u003c/div\u003e\n\u003c/div\u003e\n\u003cdiv class=\"gridtable\"\u003e\n \u003cdiv class=\"colspec\" align=\"left\"\u003e\u0026nbsp;\u003c/div\u003e\n \u003cdiv class=\"colspec\" align=\"char\"\u003e\u0026nbsp;\u003c/div\u003e\n \u003cdiv class=\"colspec\" align=\"left\"\u003e\u0026nbsp;\u003c/div\u003e\n \u003ctable id=\"Tab6\" border=\"1\"\u003e\n \u003ccaption\u003e\n \u003cdiv class=\"CaptionNumber\"\u003eTable 4\u003c/div\u003e\n \u003cdiv class=\"CaptionContent\"\u003e\n \u003cp\u003eDescriptive Statistics\u003c/p\u003e\n \u003c/div\u003e\n \u003c/caption\u003e\n \u003cthead\u003e\n \u003ctr\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003eDifficulty\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003eFrequency\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003ePercentage\u003c/p\u003e\n \u003c/th\u003e\n \u003c/tr\u003e\n \u003c/thead\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eLack of training and knowledge\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e108\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e33.90%\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eLack of infrastructure (limited access/cost)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e56\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e17.60%\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eTechnical issues\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e44\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e13.80%\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003ePoor medical knowledge\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e24\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e7.50%\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eNo difficulty reported\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e10\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e3.10%\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eMultiple challenges combined\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e77\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e24.10%\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eTotal\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e319\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e100%\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n \u003c/table\u003e\n\u003c/div\u003e"},{"header":"Discussion","content":"\u003cp\u003eNumerous research studies on the use of AI in healthcare have been conducted; however, the majorities of them were based only on radiology and were conducted on medical students and paramedic staff. This makes it necessary for us to conduct this study on the usage of AI by physicians at Peshawar's tertiary care institutions.\u003c/p\u003e\u003cp\u003eThe study highlights Peshawar's tertiary care facilities' understanding, utilization, and barriers to applying artificial intelligence into clinical operations. The findings, however, indicated a significant contradiction: while a considerable proportion of physicians (91.9%) were aware of AI, just 40.6% employed it for patient treatment. This conclusion is consistent with another poll of Khyber Pakhtunkhwa doctors, which found that 91.2% are familiar with artificial intelligence.[\u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e] This disparity between high understanding and low usage demonstrates that technology does not necessarily transfer into application, especially in situations with limited resources.\u003c/p\u003e\u003cp\u003eEven while many AI applications have already been deployed in affluent nations, this contrasts sharply with situations with limited resources, where AI applications are still in their early stages. [\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e]. Our study on patient care revealed that, whereas many clinicians were aware of AI, only 40.6% of them reported using it to treat patients.[\u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e] Similarly, a survey conducted in Peshawar found that just 11.3% of respondents used AI for practical applications, while admitting its potential to simplify clinical duties.[\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e] Our study was conducted with a large number of early career professionals such as house officers and trainee medical officers, indicating that despite being increasingly technologically exposed, practical integration remains limited.\u003c/p\u003e\u003cp\u003eEven though education is the first step, it is insufficient to drive integration, as evidenced by the significant but weakly positive relationship found by exploratory factor analysis and correlation analysis between awareness of AI and its use in clinical practice (r\u0026thinsp;=\u0026thinsp;0.151, p\u0026thinsp;=\u0026thinsp;0.013). This is consistent with research that highlights the importance of perceived usefulness in technology, workflow compatibility, and institutional support.[\u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e, \u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e23\u003c/span\u003e] This discrepancy can be further explained by the high percentage of early career professionals\u0026mdash;34.5% of the house office in our study\u0026mdash;who may not have the independence to deploy AI technologies without senior support or infrastructure.\u003c/p\u003e\u003cp\u003eLack of experience and training was the most often mentioned obstacle to adopting AI for medical care, accounting for 33.9% of responses. These findings are consistent with global research, which indicates that inadequate training remains a concern, particularly in resource-rich places.[\u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e23\u003c/span\u003e] These findings are consistent with global research, which indicates that inadequate training remains a concern, particularly in resource-rich places.[31] It's worth noting that technical concerns (13.8%) and a lack of medical experience (7.5%) were mentioned less frequently, indicating that physicians value developing their skills above instrument reliability. These problems, when considered together, underscore the need for a range of interventions that target both institutional preparation and individual skills.\u003c/p\u003e\u003cp\u003eIt was fascinating to see that just 3.1% of respondents had no difficulty using AI, underscoring the relevance of these concerns. Moderate sampling adequacy (KMO\u0026thinsp;=\u0026thinsp;0.500) supports the need for bigger studies with more diverse samples to give deeper insights into the challenges faced by various doctors.\u003c/p\u003e\u003cp\u003eThe findings suggest that AI education and infrastructure development should be a primary focus for policymakers and healthcare institutions. It should be possible for many experts to participate in training sessions and seminars in order to bridge the knowledge and practice gap. Addressing structural concerns such as high pricing, a lack of technical assistance, and uneven access to AI technology can also help increase adoption rates.\u003c/p\u003e\u003cp\u003e The study's cross-sectional design limits its ability to determine causality; the study was limited to Peshawar's tertiary care hospitals, so the findings are not generalizable; the data was self-reported, the study was limited to Peshawar's tertiary care hospitals, so the findings are not generalizable; the data was self-reported, limiting detailed insights; and some barriers to AI adoption may have gone unnoticed. Future study employing longitudinal designs, qualitative approaches, and a broader geographic scope will give a more in-depth knowledge of these components.\u003c/p\u003e"},{"header":"Conclusion","content":"\u003cp\u003eThe study shows that there is a gap between the use of AI in patient care and its awareness among the doctors of tertiary care hospitals in Peshawar. Even though a lot of doctors recognize the importance of AI only a few use it in clinical practice for patient care, this may be because of barriers like lack of infrastructure and training. The link shows that knowledge alone isn\u0026rsquo;t enough, institutional support and practical training are important too. Thus it is important for medical institutions to invest in training and technology to improve clinical practice. Future research should examine the effect of various barriers on the use of AI in patient care, and training outcomes, and explore the role of AI in multiple specialties.\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eAcknowledgement:\u0026nbsp;\u003c/strong\u003eWe express a deep gratitude to Sir Dr. Rizwan ul Haq for their guidance and continuous support. We are also thankful to the research committee of Jinnah Medical College for their approval, constructive feedback, and their support in facilitating this study.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eDisclaimer:\u0026nbsp;\u003c/strong\u003eThe authors declare no disclaimer regarding this study.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eConflict of interest:\u003c/strong\u003e The authors disclose no conflicts of interest. All opinions are those of the writers and are not affected by any personal or financial affiliations.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eFunding disclosure:\u003c/strong\u003e The study was done without any funding assistance from an outside source. No money or support has been provided by any organization, institute, or corporation.\u003cbr\u003e\u0026nbsp;\u003c/p\u003e\u003ch2\u003eAuthor Contribution\u003c/h2\u003e\u003cp\u003eA.F., A.R. and H.U. wrote the main manuscrript, A.F. and A.R. also did the data compilation and Analysis , while Q.Z. and A.Q collected the required data for the manuscript while R.H. supervised the whole research.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\u003cli\u003e\u003cspan\u003eAhmed, Z., et al., \u003cem\u003eKnowledge, attitude, and practice of artificial intelligence among doctors and medical students in Pakistan: A cross-sectional online survey\u003c/em\u003e. Ann Med Surg (Lond), 2022. 76: p. 103493.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eRigla, M., et al., \u003cem\u003eArtificial Intelligence Methodologies and Their Application to Diabetes\u003c/em\u003e. J Diabetes Sci Technol, 2018. 12(2): p. 303\u0026ndash;310.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eSadiq, F., et al., \u003cem\u003eKnowledge, Attitude, and Practice (KAP) Regarding the Use of Artificial Intelligence in Hospital Settings in Mardan, Khyber Pakhtunkhwa, Pakistan\u003c/em\u003e. Cureus, 2024. 16(12): p. e75355.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eXie, Y., Y. Zhai, and G. Lu, \u003cem\u003eEvolution of artificial intelligence in healthcare: a 30-year bibliometric study\u003c/em\u003e. Frontiers in Medicine, 2025. Volume 11\u0026ndash;2024.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eAlwadani, F.A.S., et al., \u003cem\u003eAttitude and Understanding of Artificial Intelligence Among Saudi Medical Students: An Online Cross-Sectional Study\u003c/em\u003e. J Multidiscip Healthc, 2024. 17: p. 1887\u0026ndash;1899.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eBhagat, S.V. and D. Kanyal, \u003cem\u003eNavigating the Future: The Transformative Impact of Artificial Intelligence on Hospital Management- A Comprehensive Review\u003c/em\u003e. Cureus, 2024. 16(2): p. e54518.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eChen, M. and M. Decary, \u003cem\u003eArtificial intelligence in healthcare: An essential guide for health leaders\u003c/em\u003e. Healthc Manage Forum, 2020. 33(1): p. 10\u0026ndash;18.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eMaleki Varnosfaderani, S. and M. Forouzanfar, \u003cem\u003eThe Role of AI in Hospitals and Clinics: Transforming Healthcare in the 21st Century\u003c/em\u003e. Bioengineering (Basel), 2024. 11(4).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eGiordano, C., et al., \u003cem\u003eAccessing Artificial Intelligence for Clinical Decision-Making\u003c/em\u003e. Frontiers in Digital Health, 2021. Volume 3\u0026ndash;2021.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eBrian, W., et al., \u003cem\u003eArtificial intelligence (AI) and global health: how can AI contribute to health in resource-poor settings?\u003c/em\u003e BMJ Global Health, 2018. 3(4): p. e000798.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eFei, J., et al., \u003cem\u003eArtificial intelligence in healthcare: past, present and future\u003c/em\u003e. Stroke and Vascular Neurology, 2017. 2(4).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eLi, D., K. Kulasegaram, and B.D. Hodges, \u003cem\u003eWhy We Needn't Fear the Machines: Opportunities for Medicine in a Machine Learning World\u003c/em\u003e. Acad Med, 2019. 94(5): p. 623\u0026ndash;625.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eBoillat, T., F.A. Nawaz, and H. Rivas, \u003cem\u003eReadiness to Embrace Artificial Intelligence Among Medical Doctors and Students: Questionnaire-Based Study\u003c/em\u003e. JMIR Med Educ, 2022. 8(2): p. e34973.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eBoutayeb, A., \u003cem\u003eThe double burden of communicable and non-communicable diseases in developing countries\u003c/em\u003e. Trans R Soc Trop Med Hyg, 2006. 100(3): p. 191\u0026ndash;9.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eLasrado, N., et al., \u003cem\u003eAttenuated strain of CVB3 with a mutation in the CAR-interacting region protects against both myocarditis and pancreatitis\u003c/em\u003e. Sci Rep, 2021. 11(1): p. 12432.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eTufail, A., et al., \u003cem\u003eAutomated Diabetic Retinopathy Image Assessment Software: Diagnostic Accuracy and Cost-Effectiveness Compared with Human Graders\u003c/em\u003e. Ophthalmology, 2017. 124(3): p. 343\u0026ndash;351.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003ePatoli, A.Q., \u003cem\u003eRole of syndromic management using dynamic machine learning in future of e-Health in Pakistan\u003c/em\u003e. Stud Health Technol Inform, 2007. 129(Pt 1): p. 601\u0026ndash;4.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eMalik, M., A.F. Kazi, and A. Hussain, \u003cem\u003eAdoption of health technologies for effective health information system: Need of the hour for Pakistan\u003c/em\u003e. PLoS One, 2021. 16(10): p. e0258081.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eSajjad, W., et al., \u003cem\u003eKnowledge, attitude, and practices regarding use of artificial intelligence for medical writings among doctors of Khyber Pakhtunkhwa, Pakistan: a cross-sectional study\u003c/em\u003e. Ann Med Surg (Lond), 2025. 87(3): p. 1190\u0026ndash;1199.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eMatuchansky, C., \u003cem\u003eDeep medicine, artificial intelligence, and the practising clinician\u003c/em\u003e. Lancet, 2019. 394(10200): p. 736.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eSajjad, W., et al., \u003cem\u003eKnowledge, attitude, and practices regarding use of artificial intelligence for medical writings among doctors of Khyber Pakhtunkhwa, Pakistan: a cross-sectional study\u003c/em\u003e. Annals of Medicine and Surgery, 2025. 87(3): p. 1190\u0026ndash;1199.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eJeilani, A. and S. Abubakar, \u003cem\u003ePerceived institutional support and its effects on student perceptions of AI learning in higher education: the role of mediating perceived learning outcomes and moderating technology self-efficacy\u003c/em\u003e. Frontiers in Education, 2025. Volume 10\u0026ndash;2025.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eAl-Abdullatif, A.M., \u003cem\u003eModeling Teachers\u0026rsquo; Acceptance of Generative Artificial Intelligence Use in Higher Education: The Role of AI Literacy, Intelligent TPACK, and Perceived Trust\u003c/em\u003e. Education Sciences, 2024. 14(11): p. 1209.\u003c/span\u003e\u003c/li\u003e\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":true,"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":"","lastPublishedDoi":"10.21203/rs.3.rs-7290827/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-7290827/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"Introduction: Artificial intelligence is strengthening modern healthcare by assisting with diagnosis, increasing patient monitoring, and lowering physician burden. Despite the fact that industrialized nations have begun to incorporate it into their healthcare systems, developing nations such as Pakistan continue to lag behind.\nObjective: This study assesses physicians' opinions and understanding about using AI in patient care at Peshawar's tertiary care institutions.\nMethods: This descriptive cross-sectional study was carried out between April and August of 2024 using a self-administered, closed-ended, structured questionnaire. A total of 355 people from a range of professional backgrounds participated in the study. According to the results, just 40.6% of the doctors actually employed AI, despite 91.1% of them being aware of its usage.\nResults: The results of an exploratory factor analysis showed a moderate but statistically significant correlation between AI knowledge and use (r=0.15, p=0.013). The data's eligibility for factor analysis was validated by the Bartlett's Test of Sphericity (p = 0.006) and the significant Kaiser-Meyer-Olkin score (0.500). Fisher's Exact test emphasized the importance of knowledge and implementation.\nConclusion: Despite widespread knowledge, there is little actual use of AI due to obstacles including infrastructure, support, and training. These results demonstrate how urgently Pakistan needs organized AI infrastructure and training to support the growth of its healthcare sector. Improving doctors' proficiency and trust in AI might close the knowledge gap and lead to better patient care.","manuscriptTitle":"Utilization of AI by the doctors for patient care in tertiary care hospital Peshawar","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-08-13 08:46:34","doi":"10.21203/rs.3.rs-7290827/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":"5e5373a6-0ff7-4558-8f96-967bac24cf05","owner":[],"postedDate":"August 13th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[],"tags":[],"updatedAt":"2025-08-27T05:23:37+00:00","versionOfRecord":[],"versionCreatedAt":"2025-08-13 08:46:34","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-7290827","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-7290827","identity":"rs-7290827","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}