Consistency analysis of AI cell recognition results between non anticoagulant and anticoagulant marrow smear after Wright's staining

Consistency analysis of AI cell recognition results between non anticoagulant and anticoagulant marrow smear after Wright's staining | 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 Article Consistency analysis of AI cell recognition results between non anticoagulant and anticoagulant marrow smear after Wright's staining Siheng Liu, Cenxia Ran, Jia Li, Wucheng Yang, Shuiqing Liu, Cheng Zhang, and 2 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-6195939/v1 This work is licensed under a CC BY 4.0 License Status: Under Review Version 1 posted 15 You are reading this latest preprint version Abstract The integration of artificial intelligence with bone marrow cytology represents a significant trend in the application of AI image recognition technology within the medical sector. Despite the current high accuracy of AI in cell identification, there remains a clinical need for fully automated AI cell recognition equipment that spans from sample processing to result generation. The source of the sample's origin as an influencing factor on the final results is a crucial consideration in equipment design. In this research, patient bone marrow fluid samples were processed into various types of anticoagulated bone marrow smears—EDTAK anticoagulated, sodium citrate anticoagulated, heparin lithium anticoagulated, and sodium citrate anticoagulated—and non-anticoagulated, and subsequently analyzed by AI devices. The findings revealed a significant lack of consistency in cell classification ratios and the total number of cells recognized between anticoagulated and non-anticoagulated bone marrow smear samples. This aspect must be taken into account when designing a fully automated AI-based bone marrow cell recognition device. Biological sciences/Computational biology and bioinformatics/Image processing Health sciences/Oncology/Cancer/Haematological cancer Figures Figure 1 Figure 2 Introduction The Wright’s staining method is a well-established and extensively applied technique for cell smear staining, utilized in both clinical diagnosis and scientific research. The dye eosin and methylene blue in Wright's reagent bind to NH3+ and COO- in the nucleus and cytoplasm, respectively, to stain the nucleus and cytoplasm. By identifying the characteristics of the stained nucleus and cytoplasm, people can identify and classify cells. Artificial intelligence image recognition is a breakthrough point for AI intervention in the medical field, and its application in cell recognition is a very important direction. Currently, many institutions have conducted research on this [1,2]. One of the medical technology foundations for AI cell recognition is Wright's staining. People teach AI through a large number of training cells after Wright's staining, making AI have a satisfactory accuracy rate for cell recognition [3]. One of the advantages of AI cell recognition over skilled technicians in cell classification is that it does not have mental and physical fatigue and can work for long periods of time with high throughput. The fully automated AI cell recognition work platform can fully leverage this advantage. Currently, there are few finished products on the market for fully automated AI cell recognition work platforms, and most are still in the research and design stage. At the beginning of designing a fully automated AI cell recognition work platform, there are two technical routes: samples are either manually prepared bone marrow smears or bone marrow smears prepared from anticoagulant tubes. Each of these two technical routes has its own advantages. For example, using manually prepared bone marrow smears as specimens is more closely related to traditional techniques and therefore easier to be recognized, while using anticoagulant tubes as specimens is undoubtedly more conducive to full-process automation. However, AI cell recognition training has been based on manually prepared bone marrow smears for a long time [4]. There is currently no research report on whether using anticoagulant bone marrow specimens to prepare smears will affect the accuracy rate of AI cell recognition. This article studies whether staining bone marrow smears after anticoagulation will affect the accuracy rate of AI cell recognition. Materials and Methods Basic Case Information and Sample Collection We collected 65 cases (including 43 normal cases and 22 acute leukemia cases) from our department from May 2024 to February 2025, including 43 males and 22 females, aged from 11 to 83 years. The sample collection method involved bone marrow aspiration from the patient's posterior superior iliac spine, obtaining 0.2 ml of bone marrow fluid for a manual smear, followed by another 1ml extraction from the same site into purple, blue, green, and black anticoagulant tubes [with anticoagulants: purple-EDTAK 2 , blue-sodium citrate (1:9), green-lithium heparin, and black-sodium citrate], each containing about 0.2 ml. The bone marrow fluid and anticoagulants were thoroughly mixed and then left to stand for storage. Within 2 hours, a micropipette was used to transfer 5 ul of the mixture onto a slide, which was then processed into a bone marrow smear using an automated smear machine, maintaining a constant angle and speed. Each patient's sample included five smears: one non-anticoagulated and four with anticoagulation, all stained with Wright's stain under identical conditions. Using AI platforms to identify cells and classify them Introduction to AI Platform The instrument used in this research is an AI cell recognition platform developed by our center, which consists of hardware and software components. The hardware part mainly consists of an automatic scanner and a computer. The scanner is equipped with a microscope unit with a 40x objective lens (Plan N 40 ×/0.65 FN22, resolution 0.42 μ m, Olympus, Japan), a 100x oil lens (Plan N 100 ×/1.25 FN22, resolution 0.22 μ m), and a 4000 × 3000 pixel camera (E3ISPM12000KPA, 12MP 1/1.7 inch (7.40 × 5.55) Sony Exmor CMOC sensor (ToupCam, China). The hardware environment of the computer is: Intel ® Core(TM) i9-10900X ， Nvidia GeForce RTX 3080 and ADATA DDR4 192GB. The software component is an AI automation system called Morphogo, which is based on convolutional neural networks and uses EfficientNet as its backbone to automatically extract morphological features from preprocessed cell images. The system has now trained on more than 2.8 million bone marrow nucleated cells and used 385,207 bone marrow cell images as a validation dataset. The results showed that compared with pathologists, the system had an accuracy rate of 95.55–99.98% for the classification of bone marrow nucleated cells, with an average of 99.01%. In addition, the consistency between the AI platform classification and the pathologist classification of granulocytes, red blood cells, lymphocytes, monocytes, and plasma cells was compared using the ICC method, and the results showed a high degree of consistency between the two (ICC ≥ 0.818, P < 0.01) [5]. AI platform cell classification and observation indicators According to whether or not anticoagulants are added to the samples and the types of anticoagulants added, all samples are divided into non anticoagulant group(NG), purple headed tube group(PG), green headed tube group(GG), blue headed tube group(BluG), and black headed tube group(BlaG). After each patient's 5 bone marrow smears are stained with Wright's staining, they are sequentially placed in the preview device. Senior experts select the region of interest (ROI) at the junction of the body and tail for staining and distribution based on the specific situation of the smears, and set the number of recognized and classified cells to 500. Then, all bone marrow smears are placed in the AI platform for automatic cell recognition and classification. The time required for each smear is approximately 5-10 minutes depending on the number of cells. The observation indicators of this study include: 1. visual comparison of non anticoagulant bone marrow smears and four types of anticoagulant bone marrow smears; 2. Is there a statistically significant difference in the total number of effective and fragmented cells automatically recognized by AI in the ROI area between non anticoagulant bone marrow smears and four types of anticoagulant bone marrow smears? 3. Common types of cells in the bone marrow, including myeloblast, promyelocytes, neutrophilic myelocyte , neutrophilic metamyelocyte , band neutrophil, segmented neutrophil , eosinophilia , early erythroblast , intermediate erythroblast, late erythroblast, mature lymphocyte, monocyte, and plasma cells, were subjected to consistency analysis of AI automatic recognition and classification results in the ROI area of non anticoagulant bone marrow smears and four types of anticoagulant bone marrow smears; 4. Consistency analysis of non anticoagulant bone marrow smears and four kinds of anticoagulant bone marrow smears identified by artificial intelligence with pathologists in the diagnosis of acute leukemia. Statistical analysis Perform statistical analysis using SPSS 22 and Excel 2019 software. The coefficient of variation of the total number of AI automatically recognized cells and the total number of fragmented cells in the ROI area of non anticoagulant bone marrow smears and four types of anticoagulant bone marrow smears were calculated using Excel 2019 software. The consistency analysis of AI classification ratios between non anticoagulant bone marrow smears and four types of anticoagulant bone marrow smears was conducted using the intra group correlation coefficient (ICC) method. The consistency of non anticoagulant bone marrow smears and four kinds of anticoagulant bone marrow smears identified by AI and pathologists in the diagnosis of hematological tumors was analyzed by kappa consistency test Study approval The use of human samples in this study has been approved by the Ethics Committee of Xinqiao Hospital affiliated to the Army Medical University of the People's Republic of China. All cases were sampled with written informed consent. All of the experiments using human specimens was performed in accordance with the Declaration of Helsinki and the relevant guidelines/regulations. Result Comparison of the Visual Appearance of Non-Anticoagulated Bone Marrow Smears and Four Types of Anticoagulated Bone Marrow Smears Under identical standard conditions with Wright staining, non-anticoagulated bone marrow smears demonstrated moderate blood film length, clear bone marrow particles, and well-defined head, body, and tail. Conversely, the four types of anticoagulated bone marrow smears exhibited either excessively long or short blood film lengths, reduced bone marrow particles, and indistinct head, body, and tail. The visual appearance of non-anticoagulated and four types of anticoagulated bone marrow smears is depicted in Fig. 1 . AI automatically identifies the variation coefficients for the total and fragmented cell counts After setting the AI cell recognition platform to identify 500 nucleated cells, the platform automatically scans and recognizes a preset number of nucleated cells within the ROI area. It also marks the scanned fragmented cells separately and excludes them from the total recognized cells. Depending on whether the sample contains anticoagulants or a specific type, scanning may encounter an excessive number of fragmented cells in the ROI area, failing to meet the preset number of identifiable cells. The proportions of samples where the AI recognized 500 cells within the ROI area were 95.4% for the non-anticoagulant group, 95.4% for the purple-headed tube group, 69.2% for the blue-headed tube group, 40.0% for the green-headed tube group, and 69.2% for the black-headed tube group. The total and fragmented cell counts automatically recognized by AI in the ROI area of non-anticoagulated bone marrow smears and those treated with four different anticoagulants are detailed in Table 1. The cell integrity of non-anticoagulant bone marrow smear and four types of anticoagulant bone marrow smear in the ROI area is shown in Fig. 2 . Tabel 1. AI automatic recognition of total number of cells and total number of fragmented cells in the ROI area of non anticoagulant bone marrow smears and four types of anticoagulant bone marrow smears the total number of cells of AI automatically recognizes the total number of broken cells of AI automatically recognizes Group \(\:\stackrel{-}{x}\pm\:SD\) CV, % Group \(\:\stackrel{-}{x}\pm\:SD\) CV, % NG 504.3 ± 12.5 2.5 NG 275.6 ± 231.3 83.9 PG 498.3 ± 26.9 5.4 PG 246.4 ± 190.0 77.1 BluG 440.2 ± 119.9 27.2 BluG 370.8 ± 300.6 81.1 GG 332.8 ± 173.1 52.1 GG 556.2 ± 307.8 55.3 BlaG 440.8 ± 117.0 26.5 BlaG 438.9 ± 434.4 98.9 Consistency Analysis of AI-based Recognition Between Non-Anticoagulated Bone Marrow Smears and Four Anticoagulated Variants Calculate the intraclass correlation coefficient (ICC) of the results of non anticoagulant bone marrow smears and four types of anticoagulant bone marrow smears recognized by the AI platform for all patients: 1. NG’s bone marrow smears compared with PG’s bone marrow smears showed significant consistency in 9 items (9/13, 69.2%), including 4 items with poor consistency ( promyelocytes, neutrophil myelocyte ,neutrophil metamyelocyte, intermediate erythroblast ), 4 items were of moderate consistency (myeloblast ,band neutrophil, segmented neutrophil, mature lymphocyte ), with 1 item having high consistency (late erythroblast ); 2. NG’s bone marrow smear vs. BluG’s bone marrow smear: 8 items (8/13, 61.5%) demonstrated significant consistency, with 4 item showing poor consistency (neutrophilic myelocyte, segmented neutrophil, eosinophilia, intermediate erythroblast ) and 4 items showing moderate consistency (myeloblast, band neutrophil, late erythroblast, mature lymphocyte ) ; 3. NG’s bone marrow smear vs. GG’s bone marrow smear: 8 items (8/13 ,61.5%) showed significant consistency, with 6 items having poor consistency (promyelocytes, neutrophilic myelocyte, neutrophilic metamyelocyte, segmented neutrophil, late erythroblast, mature lymphocyte), and 2 items showing moderate consistency (band neutrophil, eosinophilia ) ; 4. NG’s bone marrow smear vs. BlaG’s bone marrow smear: 9 out of 13 items (9/13, 69.2%) showed significant consistency, among which 4 had poor consistency (promyelocytes, neutrophilic myelocyte, neutrophilic metamyelocyte, intermediate erythroblast ), with 4 items showing moderate consistency (band neutrophil, segmented neutrophil, late erythroblast, mature lymphocyte ), and 1 item having high consistency (myeloblast ). The specific ICC for the results of non-anticoagulated bone marrow smears and four types of anticoagulated bone marrow smears, as identified by the AI platform, are detailed in Table 2 for all patients. Table 2 Intraclass correlation coefficients for results of non-anticoagulated and four anticoagulated bone marrow smears identified by the AI platform BlaG BluG GG PG NG myeloblast ICC 0.771 0.703 0.126 0.711 95%CI 0.650 ~ 0.854 0.555 ~ 0.808 -0.075 ~ 0.332 0.565 ~ 0.813 P 0.000 0.000 0.104 0.000 promyelocytes ICC 0.188 0.108 0.221 0.186 95%CI -0.031 ~ 0.397 -0.102 ~ 0.320 -0.002 ~ 0.429 -0.033 ~ 0.396 P 0.040 0.159 0.021 0.043 neutrophilic myelocyte ICC 0.201 0.213 0.161 0.355 95%CI -0.036 ~ 0.421 -0.021 ~ 0.429 -0.049 ~ 0.369 0.099 ~ 0.562 P 0.011 0.011 0.040 0.000 neutrophilic metamyelocyte ICC 0.256 0.182 0.238 0.317 95%CI 0.029 ~ 0.62 -0.049 ~ 0.399 0.008 ~ 0.448 0.090 ~ 0.516 P 0.011 0.061 0.021 0.002 band neutrophil ICC 0.525 0.641 0.402 0.642 95%CI 0.280 ~ 0.696 0.236 ~ 0.818 0.175 ~ 0.588 0.184 ~ 0.828 P 0.000 0.000 0.000 0.000 segmented neutrophil ICC 0.403 0.327 0.329 0.449 95%CI 0.153 ~ 0.598 0.086 ~ 0.530 0.094 ~ 0.529 0.184 ~ 0.642 P 0.000 0.001 0.004 0.000 eosinophilia ICC 0.195 0.275 0.495 0.144 95%CI -0.053 ~ 0.419 0.032 ~ 0.486 0.287 ~ 0.658 -0.074 ~ 0.357 P 0.061 0.014 0.000 0.097 early erythroblast ICC 0.006 0.004 0.111 -0.005 95%CI -0.135 ~ 0.174 -0.145 ~ 0.179 -0.099 ~ 0.323 -0.151 ~ 0.168 P 0.471 0.481 0.152 0.525 intermediate erythroblast ICC 0.320 0.282 0.173 0.398 95%CI 0.026 ~ 0.551 -0.010 ~ 0.520 -0.063 ~ 0.3934 0.147 ~ 0.595 P 0.000 0.001 0.076 0.000 late erythroblast ICC 0.579 0.681 0.185 0.780 95%CI 0.272 ~ 0.756 0.439 ~ 0.815 -0.037 ~ 0.396 0.664 ~ 0.860 P 0.000 0.000 0.048 0.000 monocyte ICC 0.130 0.125 0.078 0.073 95%CI -0.100 ~ 0.353 -0.104 ~ 0.347 -0.170 ~ 0.315 -0.135 ~ 0.287 P 0.135 0.144 0.269 0.252 mature lymphocyte ICC 0.435 0.511 0.244 0.594 95%CI 0.220 ~ 0.611 0.302 ~ 0.672 0.009 ~ 0.456 0.411 ~ 0.738 P 0.000 0.000 0.007 0.000 plasma cells ICC 0.007 0.043 -0.067 -0.011 95%CI -0.208 ~ 0.231 -0.172 ~ 0.263 -0.281 ~ 0.162 -0.224 ~ 0.213 P 0.476 0.353 0.722 0.539 Correlation analysis of AI recognition outcomes for non-anticoagulated and four anticoagulated bone marrow smears The Spearman correlation coefficient for the results obtained from the AI platform's identification of non-anticoagulated and four anticoagulated bone marrow smears in all patients : 1. NG’s bone marrow smear vs. PG’s : 10 items (10/13, 76.9%) demonstrated significant correlation, 3 of which showed weak correlation(neutrophilic myelocyte, eosinophilia, monocyte), and 4 items showed moderately correlation༈neutrophilic metamyelocyte ,segmented neutrophil, intermediate erythroblast, mature lymphocyte༉, and 3 item having high correlation (myeloblast, band neutrophil, late erythroblast ); 2. NG’s bone marrow smear vs. BluG’s : 11 items (11/13 ,84.6%) demonstrated significant correlation, 4 of which showed weak correlation (promyelocytes, neutrophilic myelocyte, neutrophilic metamyelocyte, monocyte ), and 5 items showed moderately correlation ( myeloblast, segmented neutrophil, eosinophilia, intermediate erythroblast, mature lymphocyte ), and 2 item having high correlation (band neutrophil, late erythroblast); 3. NG’s bone marrow smear vs. GG’s : 6 items (6/13 ,46.2%) demonstrated significant correlation, 3 of which showed weak correlation (band neutrophil, segmented neutrophil, monocyte), and 3 items showed moderately correlation (myeloblast, neutrophilic metamyelocyte, eosinophilia); 4. NG’s bone marrow smear vs. BlaG’s : 10 items (10/13 ,76.9%) demonstrated significant correlation, 4 of which showed weak correlation (neutrophilic myelocyte, neutrophilic metamyelocyte, eosinophilia, early erythroblast), and 4 items showed moderately correlation (band neutrophil, segmented neutrophil, intermediate erythroblast, mature lymphocyte), and 2 item having high correlation (myeloblast, late erythroblast ). The Spearman correlation analysis details for the results of non-anticoagulated and four anticoagulated bone marrow smears, identified by the AI platform, are presented in Table 3 for all patients. Table 3 Spearman Correlation Analysis for results of non-anticoagulated and four anticoagulated bone marrow smears identified by the AI platform BlaG BluG GG PG NG myeloblast ρ 0.741 0.623 0.537 0.725 P 0.000 0.000 0.000 0.000 promyelocytes ρ 0.029 0.328 0.138 0.233 P 0.816 0.008 0.274 0.062 neutrophilic myelocyte ρ 0.362 0.375 0.232 0.395 P 0.003 0.002 0.063 0.001 neutrophilic metamyelocyte ρ 0.348 0.346 0.409 0.518 P 0.004 0.005 0.001 0.000 band neutrophil ρ 0.542 0.730 0.392 0.764 P 0.000 0.000 0.001 0.000 segmented neutrophil ρ 0.485 0.403 0.354 0.540 P 0.000 0.001 0.004 0.000 eosinophilia ρ 0.374 0.494 0.450 0.360 P 0.002 0.000 0.000 0.003 early erythroblast ρ 0.274 0.219 0.235 0.106 P 0.027 0.079 0.059 0.401 intermediate erythroblast ρ 0.654 0.601 0.189 0.579 P 0.000 0.000 0.131 0.000 late erythroblast ρ 0.728 0.747 0.205 0.797 P 0.000 0.000 0.101 0.000 monocyte ρ 0.216 0.280 0.306 0.296 P 0.083 0.024 0.013 0.017 mature lymphocyte ρ 0.418 0.477 0.237 0.638 P 0.001 0.001 0.057 0.000 plasma cells ρ 0.125 0.162 0.007 0.131 P 0.321 0.198 0.958 0.300 Consistency analysis of non anticoagulant and four kinds of anticoagulant bone marrow smears identified by artificial intelligence with pathologists in the diagnosis of acute leukemia The diagnostic criterion for acute leukemia was defined as blast cells accounting for ≥ 20% of all nucleated cells. Kappa concordance analysis was performed between pathologists' diagnoses and AI-identified diagnosis from non-anticoagulated bone marrow smears along with four anticoagulated smear. The findings demonstrated substantial agreement between pathological assessments and the four anticoagulated groups, with the exception of the GG anticoagulation subgroup which showed statistically insignificant consistency. The results of Kappa consistency analysis are shown in Table 4 . Table 4 Kappa consistency analysis conclusion between pathologist's diagnosis and AI recognized non anticoagulant bone marrow smear and four types of anticoagulant smear diagnosis Pathologist's diagnosis κ P value NG 0.859 0.000 BlaG 0.893 0.000 BluG 0.856 0.000 GG 0.152 0.059 PG 0.893 0.000 Discussion Bone marrow smear cell morphology examination is the fundamental method for diagnosing hematological diseases. It examines changes in bone marrow cell morphology and proportions by analyzing cell morphology and proportions in the stained bone marrow smear. In practice, the International Council for Standardization in Haematology (ICSH) recommends not adding anticoagulants during aspiration to avoid compromising cell morphology [ 6 ]. However, with the recent surge in artificial intelligence, the integration of AI with medicine has emerged as a promising avenue of advancement. The application of AI in tumor diagnosis and treatment has consequently become a new research focus [ 7 , 8 ]. Furthermore, the integration of AI with bone marrow cell morphology is becoming increasingly intimate [ 9 ]. Compared to traditional manual bone marrow cell morphology, AI + bone marrow cell morphology offers advantages such as no fatigue issues and the ability to work continuously. Furthermore, given the uneven distribution of medical resources, AI may achieve higher diagnostic accuracy than human pathologists in certain regions [ 10 ]. However, currently, most bone marrow cell morphology AI workstations still use manually stained bone marrow smears, which is inefficient. Clinical laboratory personnel urgently require a fully automated bone marrow cell morphology AI workstation that covers the entire process from patient to report. The fully automated bone marrow cell morphology AI workstation can adopt two distinct technical approaches, differentiated primarily by the source of the sample. One approach begins with non-anticoagulated bone marrow fluid samples, where technicians prepare bone marrow smears either manually or using automated slide preparation machines. The other approach starts with anticoagulated bone marrow fluid samples; after the physician injects the bone marrow fluid into an anticoagulant tube, the workstation automatically produces the bone marrow smear. Both approaches offer unique advantages. In clinical practice, specimens for bone marrow cell identification are typically obtained from non-anticoagulated bone marrow fluid post-bone marrow puncture, which is then immediately smear-prepared by hand and stained using Wright's staining method. Specimens used for AI training are typically derived from these samples. Using anticoagulated bone marrow fluid for smear preparation enables batch processing and complete automation, thus significantly reducing processing time. In this study, although the sample sizes post-bone marrow puncture were similar between the non-anticoagulant group and the four anticoagulant tube groups, the bone marrow smears from the non-anticoagulant group outperformed those from the anticoagulant groups, both visually and statistically. The non-anticoagulant group not only exhibited clear differentiation between the head, body, and tail but also had a greater abundance of bone marrow particles compared to the anticoagulant groups. Bone marrow particles are a crucial component of bone marrow smears, providing valuable diagnostic evidence for diseases [ 8 ]. However, as bone marrow particles are solid, they tend to adhere to the tube wall after mixing with the anticoagulant fluid in the anticoagulant tube. This results in fewer particles being sampled, leading to scant or no bone marrow particles on the anticoagulant bone marrow smear. After setting the AI platform to recognize 500 cells, only the non-anticoagulant group and the purple-headed tube group successfully identified all 500 cells within the ROI area. In contrast, the blue-headed, green-headed, and black-headed tube groups only partially achieved this recognition, with only about half of the samples successfully. The blue-headed, green-headed, and black-headed tube groups showed higher numbers of fragmented cells within the ROI area compared to the non-anticoagulant and purple-headed tube groups. This is likely due to the addition of 0.2ml to 0.4ml of anticoagulant to the commercial anticoagulant tubes, which have a total anticoagulant volume of 2ml to 4ml. However, bone marrow punctures typically aspirate only 0.2 to 0.5ml of bone marrow fluid to avoid dilution. The small volume of bone marrow fluid injected into these commercial tubes leads to significant changes in osmotic pressure, resulting in numerous fragmented cells, which prevented the AI from accurately recognizing the set number of cells within the ROI area. Additionally, excessive anticoagulants lead to morphological changes in bone marrow smear cells [ 9 ]. These two factors combined lead to poor AI cell recognition outcomes for all four anticoagulated bone marrow smears and for the 13 common cells in the non-anticoagulated smears. Conclusions Currently, four anticoagulant tubes commonly used in clinical practice—namely, purple-EDTAK2, blue-sodium citrate (1:9), green-lithium heparin, and black-sodium citrate—demonstrate poor consistency in AI bone marrow cell recognition compared to non-anticoagulated bone marrow smears. This inconsistency may result from excessive anticoagulants causing bone marrow cell disruption or morphological changes. The initial design of the fully automated AI platform for bone marrow cell morphology should avoid the commonly used commercial anticoagulant tubes. Further research is required to determine whether tailoring anticoagulants to the volume of bone marrow aspirated will affect AI cell recognition outcomes. Declarations Competing interests The authors declare no competing interests. Author Contribution Si.L and X.P planned the experiments. C.R acquired human specimen. J.L and W.Y acquired and interpreted the data. Si.L wrote the manuscript. Sh.L checked the manuscript. All authors read and approved the final manuscript. Data Availability Data availability :All of the data in the present study is available from the corresponding author upon reasonable request References Eckardt JN, Middeke JM, Riechert S, et al . Deep learning detects acute myeloid leukemia and predicts NPM1 mutation status from bone marrow smears. Leukemia. 2022 Jan;36(1):111-118. doi: 10.1038/s41375-021-01408-w. Liu J, Yuan R, Li Y, et al. A deep learning method and device for bone marrow imaging cell detection. Ann Transl Med. 2022 Feb;10(4):208. doi: 10.21037/atm-22-486. Fu X, Fu M, Li Q, et al. Morphogo: An Automatic Bone Marrow Cell Classification System on Digital Images Analyzed by Artificial Intelligence. Acta Cytol. 2020;64(6):588-596. doi: 10.1159/000509524. Wu YY, Huang TC, Ye RH , et al. A Hematologist-Level Deep Learning Algorithm (BMSNet) for Assessing the Morphologies of Single Nuclear Balls in Bone Marrow Smears: Algorithm Development. JMIR Med Inform. 2020 Apr 8;8(4):e15963. doi: 10.2196/15963. Lv Z, Cao X, Jin X, et al. High-accuracy morphological identification of bone marrow cells using deep learning-based Morphogo system. Sci Rep. 2023 Aug 17;13(1):13364. doi: 10.1038/s41598-023-40424-x. Lee SH, Erber WN, Porwit A, et al. International Council for Standardization In Hematology. ICSH guidelines for the standardization of bone marrow specimens and reports. Int J Lab Hematol. 2008 Oct;30(5):349-64. doi: 10.1111/j.1751-553X.2008.01100.x. Zhang C, Xu J, Tang R, et al. Novel research and future prospects of artificial intelligence in cancer diagnosis and treatment. J Hematol Oncol. 2023 Nov 27;16(1):114. doi: 10.1186/s13045-023-01514-5. Hamamoto R, Koyama T, Kouno N, et al. Introducing AI to the molecular tumor board: one direction toward the establishment of precision medicine using large-scale cancer clinical and biological information. Exp Hematol Oncol. 2022 Oct 31;11(1):82. doi: 10.1186/s40164-022-00333-7. Eckardt JN, Middeke JM, Riechert S, et al. Deep learning detects acute myeloid leukemia and predicts NPM1 mutation status from bone marrow smears. Leukemia. 2022 Jan;36(1):111-118. doi: 10.1038/s41375-021-01408-w. Epub 2021 Sep 8. PMID: 34497326; PMCID: PMC8727290. Yu Z, Li J, Wen X, et al. AMLnet, A deep-learning pipeline for the differential diagnosis of acute myeloid leukemia from bone marrow smears. J Hematol Oncol. 2023 Mar 21;16(1):27. doi: 10.1186/s13045-023-01419-3. Zhang C, Zhang Y. Bone marrow particle enrichment analysis for the laboratory diagnosis of multiple myeloma: A case study. J Clin Lab Anal. 2020 Sep;34(9):e23372. doi: 10.1002/jcla.23372. Epub 2020 Jun 16. PMID: 32548852; PMCID: PMC7521219. Lee SH, van der Weyden C, Mayson E, et al . Excessive EDTA induces morphologic changes in bone marrow smears that mimic specific features of dysplasia. Int J Lab Hematol. 2013 Apr;35(2):163-9. doi: 10.1111/ijlh.12015. Epub 2012 Oct 13. Additional Declarations No competing interests reported. Cite Share Download PDF Status: Under Review Version 1 posted Editorial decision: Revision requested 10 Dec, 2025 Reviews received at journal 27 Nov, 2025 Reviews received at journal 23 Nov, 2025 Reviewers agreed at journal 21 Nov, 2025 Reviews received at journal 19 Nov, 2025 Reviewers agreed at journal 18 Nov, 2025 Reviewers agreed at journal 18 Nov, 2025 Reviewers agreed at journal 18 Nov, 2025 Reviewers agreed at journal 18 Nov, 2025 Reviewers agreed at journal 25 Mar, 2025 Reviewers invited by journal 24 Mar, 2025 Editor assigned by journal 24 Mar, 2025 Editor invited by journal 21 Mar, 2025 Submission checks completed at journal 20 Mar, 2025 First submitted to journal 20 Mar, 2025 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-6195939","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":433568180,"identity":"ac9acd30-1446-4fab-864a-dc7246b70dd8","order_by":0,"name":"Siheng Liu","email":"","orcid":"","institution":"Xinqiao Hospital, Army Medical University, Chongqing","correspondingAuthor":false,"prefix":"","firstName":"Siheng","middleName":"","lastName":"Liu","suffix":""},{"id":433568184,"identity":"d768e25b-f679-4fc6-8bef-40c2713d44b7","order_by":1,"name":"Cenxia Ran","email":"","orcid":"","institution":"Xinqiao Hospital, Army Medical University, Chongqing","correspondingAuthor":false,"prefix":"","firstName":"Cenxia","middleName":"","lastName":"Ran","suffix":""},{"id":433568185,"identity":"e5989819-b74d-433a-95e4-6e688efead37","order_by":2,"name":"Jia Li","email":"","orcid":"","institution":"Xinqiao Hospital, Army Medical University, Chongqing","correspondingAuthor":false,"prefix":"","firstName":"Jia","middleName":"","lastName":"Li","suffix":""},{"id":433568187,"identity":"677a58c1-0f4f-4846-8e4a-5e3df52fef8f","order_by":3,"name":"Wucheng Yang","email":"","orcid":"","institution":"Xinqiao Hospital, Army Medical University, Chongqing","correspondingAuthor":false,"prefix":"","firstName":"Wucheng","middleName":"","lastName":"Yang","suffix":""},{"id":433568188,"identity":"aeba776b-07b4-4a62-8892-1afb29eb845f","order_by":4,"name":"Shuiqing Liu","email":"","orcid":"","institution":"Xinqiao Hospital, Army Medical University, Chongqing","correspondingAuthor":false,"prefix":"","firstName":"Shuiqing","middleName":"","lastName":"Liu","suffix":""},{"id":433568190,"identity":"9ed67991-a82c-485d-b956-6566b0bdec72","order_by":5,"name":"Cheng Zhang","email":"","orcid":"","institution":"Xinqiao Hospital, Army Medical University, Chongqing","correspondingAuthor":false,"prefix":"","firstName":"Cheng","middleName":"","lastName":"Zhang","suffix":""},{"id":433568191,"identity":"5bbe57e0-ba40-4877-8dfc-1e30fdad3e06","order_by":6,"name":"Xi Zhang","email":"","orcid":"","institution":"Xinqiao Hospital, Army Medical University, Chongqing","correspondingAuthor":false,"prefix":"","firstName":"Xi","middleName":"","lastName":"Zhang","suffix":""},{"id":433568192,"identity":"156c70dd-4e35-4e78-9f67-49dffa4b0d3d","order_by":7,"name":"Xiangui Peng","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA80lEQVRIiWNgGAWjYPACCTl+hsMHoJwEorRYGEs2HktsIEVLReKG5jOGxGkxl0i/9pjnl0TiBrYz3x/dzDnMwM+eY8DwcwduLZY9Z8oNZ/ZJGG/nObuxOXfbYQbJnjcGjL1ncGsxON6TJvGxR0J25wyoFoMbOQbMjG14tBzmSZNI7JFg3HD/zUOwFnuCWo63H5P48ENCccOBM4wQWyQIaAH6hd1wZoOEsWTDMcPZudvSeSTOPCs42ItHCzDEnj3m+VMHisoHn3O3WcvxtydvfPATn8MYeMwYkJ3BAyIO4NYA0sL+jIHhDz4lo2AUjIJRMOIBABENW5iLzynKAAAAAElFTkSuQmCC","orcid":"","institution":"Xinqiao Hospital, Army Medical University, Chongqing","correspondingAuthor":true,"prefix":"","firstName":"Xiangui","middleName":"","lastName":"Peng","suffix":""}],"badges":[],"createdAt":"2025-03-10 13:24:19","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-6195939/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-6195939/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":79325971,"identity":"14800523-ca0e-43c3-8ede-833143be2193","added_by":"auto","created_at":"2025-03-27 05:31:35","extension":"jpg","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":38937,"visible":true,"origin":"","legend":"\u003cp\u003eThe visual appearance of non-anticoagulated and four types of anticoagulated bone marrow smears\u003c/p\u003e","description":"","filename":"1.jpg","url":"https://assets-eu.researchsquare.com/files/rs-6195939/v1/127bcbdc4b55d91e36906baa.jpg"},{"id":79325970,"identity":"558079b9-46ff-40ba-a993-622c9eae29d5","added_by":"auto","created_at":"2025-03-27 05:31:34","extension":"jpg","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":105108,"visible":true,"origin":"","legend":"\u003cp\u003eThe cell integrity of non-anticoagulant bone marrow smear and four types of anticoagulant bone marrow smear in the ROI area. The magnification is 1000×. (A) non anticoagulant group; (B) purple headed tube group; (C) blue headed tube group ;(D) green headed tube group; (E) black headed tube group\u003c/p\u003e","description":"","filename":"2.jpg","url":"https://assets-eu.researchsquare.com/files/rs-6195939/v1/10d7899baa1763c4b0571e6d.jpg"},{"id":79329593,"identity":"befca01e-7de4-4334-9f80-b4049e579ac9","added_by":"auto","created_at":"2025-03-27 06:12:56","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":1177959,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-6195939/v1/22b62de6-4ac0-4567-a859-12bb0d7ece40.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"Consistency analysis of AI cell recognition results between non anticoagulant and anticoagulant marrow smear after Wright's staining","fulltext":[{"header":"Introduction","content":"\u003cp\u003eThe Wright\u0026rsquo;s staining method is a well-established and extensively applied technique for cell smear staining, utilized in both clinical diagnosis and scientific research. The dye eosin and methylene blue in Wright\u0026apos;s reagent bind to NH3+ and COO- in the nucleus and cytoplasm, respectively, to stain the nucleus and cytoplasm. By identifying the characteristics of the stained nucleus and cytoplasm, people can identify and classify cells. Artificial intelligence image recognition is a breakthrough point for AI intervention in the medical field, and its application in cell recognition is a very important direction. Currently, many institutions have conducted research on this [1,2]. One of the medical technology foundations for AI cell recognition is Wright\u0026apos;s staining. People teach AI through a large number of training cells after Wright\u0026apos;s staining, making AI have a satisfactory accuracy rate for cell recognition [3]. One of the advantages of AI cell recognition over skilled technicians in cell classification is that it does not have mental and physical fatigue and can work for long periods of time with high throughput. The fully automated AI cell recognition work platform can fully leverage this advantage. Currently, there are few finished products on the market for fully automated AI cell recognition work platforms, and most are still in the research and design stage. At the beginning of designing a fully automated AI cell recognition work platform, there are two technical routes: samples are either manually prepared bone marrow smears or bone marrow smears prepared from anticoagulant tubes. Each of these two technical routes has its own advantages. For example, using manually prepared bone marrow smears as specimens is more closely related to traditional techniques and therefore easier to be recognized, while using anticoagulant tubes as specimens is undoubtedly more conducive to full-process automation. However, AI cell recognition training has been based on manually prepared bone marrow smears for a long time [4]. There is currently no research report on whether using anticoagulant bone marrow specimens to prepare smears will affect the accuracy rate of AI cell recognition. This article studies whether staining bone marrow smears after anticoagulation will affect the accuracy rate of AI cell recognition.\u003c/p\u003e"},{"header":"Materials and Methods","content":"\u003cp\u003e\u003cstrong\u003eBasic Case Information and Sample Collection\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eWe collected 65 cases (including 43 normal cases and 22 acute leukemia cases) from our department from May 2024 to February 2025, including 43 males and 22 females, aged from 11 to 83 years.\u003c/p\u003e\n\u003cp\u003eThe sample collection method involved bone marrow aspiration from the patient's posterior superior iliac spine, obtaining 0.2 ml of bone marrow fluid for a manual smear, followed by another 1ml extraction from the same site into purple, blue, green, and black anticoagulant tubes [with anticoagulants: purple-EDTAK\u003csub\u003e2\u003c/sub\u003e, blue-sodium citrate (1:9), green-lithium heparin, and black-sodium citrate], each containing about 0.2 ml. The bone marrow fluid and anticoagulants were thoroughly mixed and then left to stand for storage. Within 2 hours, a micropipette was used to transfer 5 ul of the mixture onto a slide, which was then processed into a bone marrow smear using an automated smear machine, maintaining a constant angle and speed. Each patient's sample included five smears: one non-anticoagulated and four with anticoagulation, all stained with Wright's stain under identical conditions.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eUsing AI platforms to identify cells and classify them\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eIntroduction to AI Platform\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe instrument used in this research is an AI cell recognition platform developed by our center, which consists of hardware and software components. The hardware part mainly consists of an automatic scanner and a computer. The scanner is equipped with a microscope unit with a 40x objective lens (Plan N 40 ×/0.65 FN22, resolution 0.42 μ m, Olympus, Japan), a 100x oil lens (Plan N 100 ×/1.25 FN22, resolution 0.22 μ m), and a 4000 × 3000 pixel camera (E3ISPM12000KPA, 12MP 1/1.7 inch (7.40 × 5.55) Sony Exmor CMOC sensor (ToupCam, China). The hardware environment of the computer is: Intel ® Core(TM) i9-10900X ， Nvidia GeForce RTX 3080 and ADATA DDR4 192GB.\u003c/p\u003e\n\u003cp\u003eThe software component is an AI automation system called Morphogo, which is based on convolutional neural networks and uses EfficientNet as its backbone to automatically extract morphological features from preprocessed cell images. The system has now trained on more than 2.8 million bone marrow nucleated cells and used 385,207 bone marrow cell images as a validation dataset. The results showed that compared with pathologists, the system had an accuracy rate of 95.55–99.98% for the classification of bone marrow nucleated cells, with an average of 99.01%. In addition, the consistency between the AI platform classification and the pathologist classification of granulocytes, red blood cells, lymphocytes, monocytes, and plasma cells was compared using the ICC method, and the results showed a high degree of consistency between the two (ICC ≥ 0.818, P \u0026lt; 0.01) [5].\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAI platform cell classification and observation indicators\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eAccording to whether or not anticoagulants are added to the samples and the types of anticoagulants added, all samples are divided into non anticoagulant group(NG),\u0026nbsp;purple headed tube group(PG), green headed tube group(GG), blue headed tube group(BluG), and black headed tube group(BlaG). After each patient's 5 bone marrow smears are stained with Wright's staining, they are sequentially placed in the preview device. Senior experts select the region of interest (ROI) at the junction of the body and tail for staining and distribution based on the specific situation of the smears, and set the number of recognized and classified cells to 500. Then, all bone marrow smears are placed in the AI platform for automatic cell recognition and classification. The time required for each smear is approximately 5-10 minutes depending on the number of cells.\u003c/p\u003e\n\u003cp\u003eThe observation indicators of this study include: 1. visual comparison of non anticoagulant bone marrow smears and four types of anticoagulant bone marrow smears; 2. Is there a statistically significant difference in the total number of effective and fragmented cells automatically recognized by AI in the ROI area between non anticoagulant bone marrow smears and four types of anticoagulant bone marrow smears? \u0026nbsp;3. Common types of cells in the bone marrow, including\u003c/p\u003e\n\u003cp\u003emyeloblast, promyelocytes, neutrophilic myelocyte , neutrophilic metamyelocyte , band neutrophil, \u0026nbsp; segmented neutrophil , eosinophilia , early erythroblast , intermediate erythroblast, late erythroblast, mature lymphocyte, monocyte, and plasma cells, were subjected to consistency analysis of AI automatic recognition and classification results in the ROI area of non anticoagulant bone marrow smears and four types of anticoagulant bone marrow smears; 4.\u0026nbsp;Consistency analysis of non anticoagulant bone marrow smears and four kinds of anticoagulant bone marrow smears identified by artificial intelligence with pathologists in the diagnosis of\u0026nbsp;acute leukemia.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eStatistical analysis\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003ePerform statistical analysis using SPSS 22 and Excel 2019 software. The coefficient of variation of the total number of AI automatically recognized cells and the total number of fragmented cells in the ROI area of non anticoagulant bone marrow smears and four types of anticoagulant bone marrow smears were calculated using Excel 2019 software. The consistency analysis of AI classification ratios between non anticoagulant bone marrow smears and four types of anticoagulant bone marrow smears was conducted using the intra group correlation coefficient (ICC) method. The consistency of non anticoagulant bone marrow smears and four kinds of anticoagulant bone marrow smears identified by AI and pathologists in the diagnosis of hematological tumors was analyzed by kappa consistency test\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eStudy approval\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe use of human samples in this study has been approved by the Ethics Committee of Xinqiao Hospital affiliated to the Army Medical University of the People's Republic of China. All cases were sampled with written informed consent. All of the experiments using human specimens was performed in accordance with the Declaration of Helsinki and the relevant guidelines/regulations.\u003c/p\u003e"},{"header":"Result","content":"\u003cp\u003e \u003cb\u003eComparison of the Visual Appearance of Non-Anticoagulated Bone Marrow Smears and Four Types of Anticoagulated Bone Marrow Smears\u003c/b\u003e \u003c/p\u003e \u003cp\u003eUnder identical standard conditions with Wright staining, non-anticoagulated bone marrow smears demonstrated moderate blood film length, clear bone marrow particles, and well-defined head, body, and tail. Conversely, the four types of anticoagulated bone marrow smears exhibited either excessively long or short blood film lengths, reduced bone marrow particles, and indistinct head, body, and tail. The visual appearance of non-anticoagulated and four types of anticoagulated bone marrow smears is depicted in Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e\n\u003ch3\u003eAI automatically identifies the variation coefficients for the total and fragmented cell counts\u003c/h3\u003e\n\u003cp\u003eAfter setting the AI cell recognition platform to identify 500 nucleated cells, the platform automatically scans and recognizes a preset number of nucleated cells within the ROI area. It also marks the scanned fragmented cells separately and excludes them from the total recognized cells. Depending on whether the sample contains anticoagulants or a specific type, scanning may encounter an excessive number of fragmented cells in the ROI area, failing to meet the preset number of identifiable cells. The proportions of samples where the AI recognized 500 cells within the ROI area were 95.4% for the non-anticoagulant group, 95.4% for the purple-headed tube group, 69.2% for the blue-headed tube group, 40.0% for the green-headed tube group, and 69.2% for the black-headed tube group. The total and fragmented cell counts automatically recognized by AI in the ROI area of non-anticoagulated bone marrow smears and those treated with four different anticoagulants are detailed in Table\u0026nbsp;1. The cell integrity of non-anticoagulant bone marrow smear and four types of anticoagulant bone marrow smear in the ROI area is shown in Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003e.\u003c/p\u003e \u003cp\u003eTabel 1. AI automatic recognition of total number of cells and total number of fragmented cells in the ROI area of non anticoagulant bone marrow smears and four types of anticoagulant bone marrow smears\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"No\" id=\"Taba\" border=\"1\"\u003e \u003ccolgroup cols=\"6\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" 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 \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colspan=\"3\" nameend=\"c3\" namest=\"c1\"\u003e \u003cp\u003ethe total number of cells of AI automatically recognizes\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colspan=\"3\" nameend=\"c6\" namest=\"c4\"\u003e \u003cp\u003ethe total number of broken cells of AI automatically recognizes\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eGroup\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e\u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:\\stackrel{-}{x}\\pm\\:SD\\)\u003c/span\u003e\u003c/span\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eCV, %\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eGroup\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e\u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:\\stackrel{-}{x}\\pm\\:SD\\)\u003c/span\u003e\u003c/span\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003eCV, %\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eNG\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e504.3\u0026thinsp;\u0026plusmn;\u0026thinsp;12.5\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e2.5\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eNG\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e275.6\u0026thinsp;\u0026plusmn;\u0026thinsp;231.3\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e83.9\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ePG\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e498.3\u0026thinsp;\u0026plusmn;\u0026thinsp;26.9\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e5.4\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003ePG\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e246.4\u0026thinsp;\u0026plusmn;\u0026thinsp;190.0\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e77.1\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eBluG\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e440.2\u0026thinsp;\u0026plusmn;\u0026thinsp;119.9\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e27.2\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eBluG\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e370.8\u0026thinsp;\u0026plusmn;\u0026thinsp;300.6\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e81.1\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eGG\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e332.8\u0026thinsp;\u0026plusmn;\u0026thinsp;173.1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e52.1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eGG\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e556.2\u0026thinsp;\u0026plusmn;\u0026thinsp;307.8\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e55.3\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eBlaG\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e440.8\u0026thinsp;\u0026plusmn;\u0026thinsp;117.0\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e26.5\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eBlaG\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e438.9\u0026thinsp;\u0026plusmn;\u0026thinsp;434.4\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e98.9\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003e \u003c/p\u003e\n\u003ch3\u003eConsistency Analysis of AI-based Recognition Between Non-Anticoagulated Bone Marrow Smears and Four Anticoagulated Variants\u003c/h3\u003e\n\u003cp\u003eCalculate the intraclass correlation coefficient (ICC) of the results of non anticoagulant bone marrow smears and four types of anticoagulant bone marrow smears recognized by the AI platform for all patients: 1. NG\u0026rsquo;s bone marrow smears compared with PG\u0026rsquo;s bone marrow smears showed significant consistency in 9 items (9/13, 69.2%), including 4 items with poor consistency ( promyelocytes, neutrophil myelocyte ,neutrophil metamyelocyte, intermediate erythroblast ), 4 items were of moderate consistency (myeloblast ,band neutrophil, segmented neutrophil, mature lymphocyte ), with 1 item having high consistency (late erythroblast ); 2. NG\u0026rsquo;s bone marrow smear vs. BluG\u0026rsquo;s bone marrow smear: 8 items (8/13, 61.5%) demonstrated significant consistency, with 4 item showing poor consistency (neutrophilic myelocyte, segmented neutrophil, eosinophilia, intermediate erythroblast ) and 4 items showing moderate consistency (myeloblast, band neutrophil, late erythroblast, mature lymphocyte ) ; 3. NG\u0026rsquo;s bone marrow smear vs. GG\u0026rsquo;s bone marrow smear: 8 items (8/13 ,61.5%) showed significant consistency, with 6 items having poor consistency (promyelocytes, neutrophilic myelocyte, neutrophilic metamyelocyte, segmented neutrophil, late erythroblast, mature lymphocyte), and 2 items showing moderate consistency (band neutrophil, eosinophilia ) ; 4. NG\u0026rsquo;s bone marrow smear vs. BlaG\u0026rsquo;s bone marrow smear: 9 out of 13 items (9/13, 69.2%) showed significant consistency, among which 4 had poor consistency (promyelocytes, neutrophilic myelocyte, neutrophilic metamyelocyte, intermediate erythroblast ), with 4 items showing moderate consistency (band neutrophil, segmented neutrophil, late erythroblast, mature lymphocyte ), and 1 item having high consistency (myeloblast ). The specific ICC for the results of non-anticoagulated bone marrow smears and four types of anticoagulated bone marrow smears, as identified by the AI platform, are detailed in Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e2\u003c/span\u003e for all patients.\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 2\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eIntraclass correlation coefficients for results of non-anticoagulated and four anticoagulated bone marrow smears identified by the AI platform\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"7\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" 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=\"char\" char=\".\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c7\" colnum=\"7\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colspan=\"3\" nameend=\"c3\" namest=\"c1\"\u003e\u0026nbsp;\u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eBlaG\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003eBluG\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c6\"\u003e \u003cp\u003eGG\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c7\"\u003e \u003cp\u003ePG\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\" morerows=\"38\" rowspan=\"39\"\u003e \u003cp\u003eNG\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\" morerows=\"2\" rowspan=\"3\"\u003e \u003cp\u003emyeloblast\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eICC\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.771\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.703\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.126\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.711\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e95%CI\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.650\u0026thinsp;~\u0026thinsp;0.854\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.555\u0026thinsp;~\u0026thinsp;0.808\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e-0.075\u0026thinsp;~\u0026thinsp;0.332\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.565\u0026thinsp;~\u0026thinsp;0.813\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eP\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.000\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.000\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.104\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.000\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\" morerows=\"2\" rowspan=\"3\"\u003e \u003cp\u003epromyelocytes\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eICC\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.188\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.108\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.221\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.186\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e95%CI\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e-0.031\u0026thinsp;~\u0026thinsp;0.397\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e-0.102\u0026thinsp;~\u0026thinsp;0.320\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e-0.002\u0026thinsp;~\u0026thinsp;0.429\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e-0.033\u0026thinsp;~\u0026thinsp;0.396\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eP\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.040\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.159\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.021\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.043\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\" morerows=\"2\" rowspan=\"3\"\u003e \u003cp\u003eneutrophilic myelocyte\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eICC\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.201\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.213\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.161\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.355\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e95%CI\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e-0.036\u0026thinsp;~\u0026thinsp;0.421\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e-0.021\u0026thinsp;~\u0026thinsp;0.429\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e-0.049\u0026thinsp;~\u0026thinsp;0.369\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.099\u0026thinsp;~\u0026thinsp;0.562\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eP\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.011\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.011\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.040\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.000\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\" morerows=\"2\" rowspan=\"3\"\u003e \u003cp\u003eneutrophilic metamyelocyte\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eICC\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.256\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.182\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.238\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.317\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e95%CI\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.029\u0026thinsp;~\u0026thinsp;0.62\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e-0.049\u0026thinsp;~\u0026thinsp;0.399\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.008\u0026thinsp;~\u0026thinsp;0.448\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.090\u0026thinsp;~\u0026thinsp;0.516\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eP\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.011\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.061\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.021\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.002\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\" morerows=\"2\" rowspan=\"3\"\u003e \u003cp\u003eband neutrophil\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eICC\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.525\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.641\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.402\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.642\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e95%CI\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.280\u0026thinsp;~\u0026thinsp;0.696\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.236\u0026thinsp;~\u0026thinsp;0.818\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.175\u0026thinsp;~\u0026thinsp;0.588\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.184\u0026thinsp;~\u0026thinsp;0.828\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eP\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.000\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.000\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.000\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.000\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\" morerows=\"2\" rowspan=\"3\"\u003e \u003cp\u003esegmented neutrophil\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eICC\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.403\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.327\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.329\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.449\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e95%CI\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.153\u0026thinsp;~\u0026thinsp;0.598\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.086\u0026thinsp;~\u0026thinsp;0.530\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.094\u0026thinsp;~\u0026thinsp;0.529\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.184\u0026thinsp;~\u0026thinsp;0.642\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eP\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.000\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.001\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.004\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.000\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\" morerows=\"2\" rowspan=\"3\"\u003e \u003cp\u003eeosinophilia\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eICC\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.195\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.275\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.495\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.144\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e95%CI\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e-0.053\u0026thinsp;~\u0026thinsp;0.419\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.032\u0026thinsp;~\u0026thinsp;0.486\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.287\u0026thinsp;~\u0026thinsp;0.658\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e-0.074\u0026thinsp;~\u0026thinsp;0.357\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eP\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.061\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.014\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.000\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.097\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\" morerows=\"2\" rowspan=\"3\"\u003e \u003cp\u003eearly erythroblast\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eICC\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.006\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.004\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.111\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e-0.005\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e95%CI\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e-0.135\u0026thinsp;~\u0026thinsp;0.174\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e-0.145\u0026thinsp;~\u0026thinsp;0.179\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e-0.099\u0026thinsp;~\u0026thinsp;0.323\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e-0.151\u0026thinsp;~\u0026thinsp;0.168\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eP\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.471\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.481\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.152\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.525\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\" morerows=\"2\" rowspan=\"3\"\u003e \u003cp\u003eintermediate erythroblast\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eICC\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.320\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.282\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.173\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.398\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e95%CI\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.026\u0026thinsp;~\u0026thinsp;0.551\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e-0.010\u0026thinsp;~\u0026thinsp;0.520\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e-0.063\u0026thinsp;~\u0026thinsp;0.3934\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.147\u0026thinsp;~\u0026thinsp;0.595\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eP\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.000\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.001\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.076\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.000\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\" morerows=\"2\" rowspan=\"3\"\u003e \u003cp\u003elate erythroblast\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eICC\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.579\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.681\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.185\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.780\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e95%CI\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.272\u0026thinsp;~\u0026thinsp;0.756\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.439\u0026thinsp;~\u0026thinsp;0.815\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e-0.037\u0026thinsp;~\u0026thinsp;0.396\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.664\u0026thinsp;~\u0026thinsp;0.860\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eP\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.000\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.000\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.048\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.000\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\" morerows=\"2\" rowspan=\"3\"\u003e \u003cp\u003emonocyte\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eICC\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.130\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.125\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.078\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.073\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e95%CI\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e-0.100\u0026thinsp;~\u0026thinsp;0.353\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e-0.104\u0026thinsp;~\u0026thinsp;0.347\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e-0.170\u0026thinsp;~\u0026thinsp;0.315\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e-0.135\u0026thinsp;~\u0026thinsp;0.287\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eP\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.135\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.144\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.269\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.252\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\" morerows=\"2\" rowspan=\"3\"\u003e \u003cp\u003emature lymphocyte\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eICC\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.435\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.511\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.244\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.594\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e95%CI\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.220\u0026thinsp;~\u0026thinsp;0.611\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.302\u0026thinsp;~\u0026thinsp;0.672\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.009\u0026thinsp;~\u0026thinsp;0.456\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.411\u0026thinsp;~\u0026thinsp;0.738\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eP\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.000\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.000\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.007\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.000\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\" morerows=\"2\" rowspan=\"3\"\u003e \u003cp\u003eplasma cells\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eICC\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.007\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.043\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e-0.067\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e-0.011\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e95%CI\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e-0.208\u0026thinsp;~\u0026thinsp;0.231\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e-0.172\u0026thinsp;~\u0026thinsp;0.263\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e-0.281\u0026thinsp;~\u0026thinsp;0.162\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e-0.224\u0026thinsp;~\u0026thinsp;0.213\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eP\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.476\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.353\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.722\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.539\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cdiv id=\"Sec8\" class=\"Section2\"\u003e \u003ch2\u003eCorrelation analysis of AI recognition outcomes for non-anticoagulated and four anticoagulated bone marrow smears\u003c/h2\u003e \u003cp\u003eThe Spearman correlation coefficient for the results obtained from the AI platform's identification of non-anticoagulated and four anticoagulated bone marrow smears in all patients : 1. NG\u0026rsquo;s bone marrow smear vs. PG\u0026rsquo;s : 10 items (10/13, 76.9%) demonstrated significant correlation, 3 of which showed weak correlation(neutrophilic myelocyte, eosinophilia, monocyte), and 4 items showed moderately correlation༈neutrophilic metamyelocyte ,segmented neutrophil, intermediate erythroblast, mature lymphocyte༉, and 3 item having high correlation (myeloblast, band neutrophil, late erythroblast ); 2. NG\u0026rsquo;s bone marrow smear vs. BluG\u0026rsquo;s : 11 items (11/13 ,84.6%) demonstrated significant correlation, 4 of which showed weak correlation (promyelocytes, neutrophilic myelocyte, neutrophilic metamyelocyte, monocyte ), and 5 items showed moderately correlation ( myeloblast, segmented neutrophil, eosinophilia, intermediate erythroblast, mature lymphocyte ), and 2 item having high correlation (band neutrophil, late erythroblast); 3. NG\u0026rsquo;s bone marrow smear vs. GG\u0026rsquo;s : 6 items (6/13 ,46.2%) demonstrated significant correlation, 3 of which showed weak correlation (band neutrophil, segmented neutrophil, monocyte), and 3 items showed moderately correlation (myeloblast, neutrophilic metamyelocyte, eosinophilia); 4. NG\u0026rsquo;s bone marrow smear vs. BlaG\u0026rsquo;s : 10 items (10/13 ,76.9%) demonstrated significant correlation, 4 of which showed weak correlation (neutrophilic myelocyte, neutrophilic metamyelocyte, eosinophilia, early erythroblast), and 4 items showed moderately correlation (band neutrophil, segmented neutrophil, intermediate erythroblast, mature lymphocyte), and 2 item having high correlation (myeloblast, late erythroblast ). The Spearman correlation analysis details for the results of non-anticoagulated and four anticoagulated bone marrow smears, identified by the AI platform, are presented in Table\u0026nbsp;\u003cspan refid=\"Tab2\" class=\"InternalRef\"\u003e3\u003c/span\u003e for all patients.\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab2\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 3\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eSpearman Correlation Analysis for results of non-anticoagulated and four anticoagulated bone marrow smears identified by the AI platform\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"7\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" 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=\"char\" char=\".\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c7\" colnum=\"7\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colspan=\"3\" nameend=\"c3\" namest=\"c1\"\u003e\u0026nbsp;\u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eBlaG\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003eBluG\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c6\"\u003e \u003cp\u003eGG\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c7\"\u003e \u003cp\u003ePG\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\" morerows=\"25\" rowspan=\"26\"\u003e \u003cp\u003eNG\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\" morerows=\"1\" rowspan=\"2\"\u003e \u003cp\u003emyeloblast\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eρ\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.741\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.623\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.537\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.725\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eP\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.000\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.000\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.000\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.000\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\" morerows=\"1\" rowspan=\"2\"\u003e \u003cp\u003epromyelocytes\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eρ\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.029\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.328\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.138\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.233\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eP\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.816\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.008\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.274\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.062\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\" morerows=\"1\" rowspan=\"2\"\u003e \u003cp\u003eneutrophilic myelocyte\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eρ\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.362\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.375\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.232\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.395\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eP\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.003\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.002\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.063\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.001\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\" morerows=\"1\" rowspan=\"2\"\u003e \u003cp\u003eneutrophilic metamyelocyte\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eρ\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.348\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.346\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.409\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.518\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eP\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.004\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.005\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.001\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.000\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\" morerows=\"1\" rowspan=\"2\"\u003e \u003cp\u003eband neutrophil\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eρ\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.542\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.730\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.392\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.764\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eP\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.000\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.000\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.001\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.000\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\" morerows=\"1\" rowspan=\"2\"\u003e \u003cp\u003esegmented neutrophil\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eρ\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.485\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.403\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.354\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.540\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eP\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.000\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.001\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.004\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.000\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\" morerows=\"1\" rowspan=\"2\"\u003e \u003cp\u003eeosinophilia\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eρ\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.374\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.494\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.450\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.360\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eP\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.002\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.000\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.000\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.003\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\" morerows=\"1\" rowspan=\"2\"\u003e \u003cp\u003eearly erythroblast\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eρ\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.274\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.219\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.235\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.106\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eP\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.027\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.079\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.059\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.401\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\" morerows=\"1\" rowspan=\"2\"\u003e \u003cp\u003eintermediate erythroblast\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eρ\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.654\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.601\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.189\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.579\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eP\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.000\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.000\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.131\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.000\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\" morerows=\"1\" rowspan=\"2\"\u003e \u003cp\u003elate erythroblast\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eρ\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.728\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.747\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.205\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.797\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eP\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.000\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.000\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.101\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.000\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\" morerows=\"1\" rowspan=\"2\"\u003e \u003cp\u003emonocyte\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eρ\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.216\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.280\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.306\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.296\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eP\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.083\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.024\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.013\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.017\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\" morerows=\"1\" rowspan=\"2\"\u003e \u003cp\u003emature lymphocyte\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eρ\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.418\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.477\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.237\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.638\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eP\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.001\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.001\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.057\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.000\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\" morerows=\"1\" rowspan=\"2\"\u003e \u003cp\u003eplasma cells\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eρ\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.125\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.162\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.007\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.131\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eP\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.321\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.198\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.958\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.300\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003e \u003cb\u003eConsistency analysis of non anticoagulant and four kinds of anticoagulant bone marrow smears identified by artificial intelligence with pathologists in the diagnosis of acute leukemia\u003c/b\u003e \u003c/p\u003e \u003cp\u003eThe diagnostic criterion for acute leukemia was defined as blast cells accounting for \u0026ge;\u0026thinsp;20% of all nucleated cells. Kappa concordance analysis was performed between pathologists' diagnoses and AI-identified diagnosis from non-anticoagulated bone marrow smears along with four anticoagulated smear. The findings demonstrated substantial agreement between pathological assessments and the four anticoagulated groups, with the exception of the GG anticoagulation subgroup which showed statistically insignificant consistency. The results of Kappa consistency analysis are shown in Table\u0026nbsp;\u003cspan refid=\"Tab3\" class=\"InternalRef\"\u003e4\u003c/span\u003e.\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab3\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 4\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eKappa consistency analysis conclusion between pathologist's diagnosis and AI recognized non anticoagulant bone marrow smear and four types of anticoagulant smear diagnosis\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"3\"\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 \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/th\u003e \u003cth align=\"left\" colspan=\"2\" nameend=\"c3\" namest=\"c2\"\u003e \u003cp\u003ePathologist's diagnosis\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eκ\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eP value\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eNG\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.859\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.000\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eBlaG\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.893\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.000\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eBluG\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.856\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.000\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eGG\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.152\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.059\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ePG\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.893\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.000\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003c/div\u003e"},{"header":"Discussion","content":"\u003cp\u003eBone marrow smear cell morphology examination is the fundamental method for diagnosing hematological diseases. It examines changes in bone marrow cell morphology and proportions by analyzing cell morphology and proportions in the stained bone marrow smear. In practice, the International Council for Standardization in Haematology (ICSH) recommends not adding anticoagulants during aspiration to avoid compromising cell morphology [\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e]. However, with the recent surge in artificial intelligence, the integration of AI with medicine has emerged as a promising avenue of advancement. The application of AI in tumor diagnosis and treatment has consequently become a new research focus [\u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e, \u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e]. Furthermore, the integration of AI with bone marrow cell morphology is becoming increasingly intimate [\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e]. Compared to traditional manual bone marrow cell morphology, AI\u0026thinsp;+\u0026thinsp;bone marrow cell morphology offers advantages such as no fatigue issues and the ability to work continuously. Furthermore, given the uneven distribution of medical resources, AI may achieve higher diagnostic accuracy than human pathologists in certain regions [\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e]. However, currently, most bone marrow cell morphology AI workstations still use manually stained bone marrow smears, which is inefficient. Clinical laboratory personnel urgently require a fully automated bone marrow cell morphology AI workstation that covers the entire process from patient to report.\u003c/p\u003e \u003cp\u003eThe fully automated bone marrow cell morphology AI workstation can adopt two distinct technical approaches, differentiated primarily by the source of the sample. One approach begins with non-anticoagulated bone marrow fluid samples, where technicians prepare bone marrow smears either manually or using automated slide preparation machines. The other approach starts with anticoagulated bone marrow fluid samples; after the physician injects the bone marrow fluid into an anticoagulant tube, the workstation automatically produces the bone marrow smear. Both approaches offer unique advantages. In clinical practice, specimens for bone marrow cell identification are typically obtained from non-anticoagulated bone marrow fluid post-bone marrow puncture, which is then immediately smear-prepared by hand and stained using Wright's staining method. Specimens used for AI training are typically derived from these samples. Using anticoagulated bone marrow fluid for smear preparation enables batch processing and complete automation, thus significantly reducing processing time.\u003c/p\u003e \u003cp\u003eIn this study, although the sample sizes post-bone marrow puncture were similar between the non-anticoagulant group and the four anticoagulant tube groups, the bone marrow smears from the non-anticoagulant group outperformed those from the anticoagulant groups, both visually and statistically. The non-anticoagulant group not only exhibited clear differentiation between the head, body, and tail but also had a greater abundance of bone marrow particles compared to the anticoagulant groups. Bone marrow particles are a crucial component of bone marrow smears, providing valuable diagnostic evidence for diseases [\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e]. However, as bone marrow particles are solid, they tend to adhere to the tube wall after mixing with the anticoagulant fluid in the anticoagulant tube. This results in fewer particles being sampled, leading to scant or no bone marrow particles on the anticoagulant bone marrow smear.\u003c/p\u003e \u003cp\u003eAfter setting the AI platform to recognize 500 cells, only the non-anticoagulant group and the purple-headed tube group successfully identified all 500 cells within the ROI area. In contrast, the blue-headed, green-headed, and black-headed tube groups only partially achieved this recognition, with only about half of the samples successfully. The blue-headed, green-headed, and black-headed tube groups showed higher numbers of fragmented cells within the ROI area compared to the non-anticoagulant and purple-headed tube groups. This is likely due to the addition of 0.2ml to 0.4ml of anticoagulant to the commercial anticoagulant tubes, which have a total anticoagulant volume of 2ml to 4ml. However, bone marrow punctures typically aspirate only 0.2 to 0.5ml of bone marrow fluid to avoid dilution. The small volume of bone marrow fluid injected into these commercial tubes leads to significant changes in osmotic pressure, resulting in numerous fragmented cells, which prevented the AI from accurately recognizing the set number of cells within the ROI area. Additionally, excessive anticoagulants lead to morphological changes in bone marrow smear cells [\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e]. These two factors combined lead to poor AI cell recognition outcomes for all four anticoagulated bone marrow smears and for the 13 common cells in the non-anticoagulated smears.\u003c/p\u003e"},{"header":"Conclusions","content":"\u003cp\u003eCurrently, four anticoagulant tubes commonly used in clinical practice\u0026mdash;namely, purple-EDTAK2, blue-sodium citrate (1:9), green-lithium heparin, and black-sodium citrate\u0026mdash;demonstrate poor consistency in AI bone marrow cell recognition compared to non-anticoagulated bone marrow smears. This inconsistency may result from excessive anticoagulants causing bone marrow cell disruption or morphological changes. The initial design of the fully automated AI platform for bone marrow cell morphology should avoid the commonly used commercial anticoagulant tubes. Further research is required to determine whether tailoring anticoagulants to the volume of bone marrow aspirated will affect AI cell recognition outcomes.\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e \u003ch2\u003eCompeting interests\u003c/h2\u003e \u003cp\u003eThe authors declare no competing interests.\u003c/p\u003e \u003c/p\u003e\u003ch2\u003eAuthor Contribution\u003c/h2\u003e\u003cp\u003eSi.L and X.P planned the experiments. C.R acquired human specimen. J.L and W.Y acquired and interpreted the data. Si.L wrote the manuscript. Sh.L checked the manuscript. All authors read and approved the final manuscript.\u003c/p\u003e\u003ch2\u003eData Availability\u003c/h2\u003e\u003cp\u003eData availability :All of the data in the present study is available from the corresponding author upon reasonable request\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n\u003cli\u003eEckardt JN, Middeke JM, Riechert S, et al . Deep learning detects acute myeloid leukemia and predicts NPM1 mutation status from bone marrow smears. Leukemia. 2022 Jan;36(1):111-118. doi: 10.1038/s41375-021-01408-w. \u003c/li\u003e\n\u003cli\u003eLiu J, Yuan R, Li Y, et al. A deep learning method and device for bone marrow imaging cell detection. Ann Transl Med. 2022 Feb;10(4):208. doi: 10.21037/atm-22-486. \u003c/li\u003e\n\u003cli\u003eFu X, Fu M, Li Q, et al. Morphogo: An Automatic Bone Marrow Cell Classification System on Digital Images Analyzed by Artificial Intelligence. Acta Cytol. 2020;64(6):588-596. doi: 10.1159/000509524. \u003c/li\u003e\n\u003cli\u003eWu YY, Huang TC, Ye RH , et al. A Hematologist-Level Deep Learning Algorithm (BMSNet) for Assessing the Morphologies of Single Nuclear Balls in Bone Marrow Smears: Algorithm Development. JMIR Med Inform. 2020 Apr 8;8(4):e15963. doi: 10.2196/15963. \u003c/li\u003e\n\u003cli\u003eLv Z, Cao X, Jin X, et al. High-accuracy morphological identification of bone marrow cells using deep learning-based Morphogo system. Sci Rep. 2023 Aug 17;13(1):13364. doi: 10.1038/s41598-023-40424-x. \u003c/li\u003e\n\u003cli\u003eLee SH, Erber WN, Porwit A, et al. International Council for Standardization In Hematology. ICSH guidelines for the standardization of bone marrow specimens and reports. Int J Lab Hematol. 2008 Oct;30(5):349-64. doi: 10.1111/j.1751-553X.2008.01100.x. \u003c/li\u003e\n\u003cli\u003eZhang C, Xu J, Tang R, et al. Novel research and future prospects of artificial intelligence in cancer diagnosis and treatment. J Hematol Oncol. 2023 Nov 27;16(1):114. doi: 10.1186/s13045-023-01514-5.\u003c/li\u003e\n\u003cli\u003eHamamoto R, Koyama T, Kouno N, et al. Introducing AI to the molecular tumor board: one direction toward the establishment of precision medicine using large-scale cancer clinical and biological information. Exp Hematol Oncol. 2022 Oct 31;11(1):82. doi: 10.1186/s40164-022-00333-7.\u003c/li\u003e\n\u003cli\u003eEckardt JN, Middeke JM, Riechert S, et al. Deep learning detects acute myeloid leukemia and predicts NPM1 mutation status from bone marrow smears. Leukemia. 2022 Jan;36(1):111-118. doi: 10.1038/s41375-021-01408-w. Epub 2021 Sep 8. PMID: 34497326; PMCID: PMC8727290.\u003c/li\u003e\n\u003cli\u003eYu Z, Li J, Wen X, et al. AMLnet, A deep-learning pipeline for the differential diagnosis of acute myeloid leukemia from bone marrow smears. J Hematol Oncol. 2023 Mar 21;16(1):27. doi: 10.1186/s13045-023-01419-3.\u003c/li\u003e\n\u003cli\u003eZhang C, Zhang Y. Bone marrow particle enrichment analysis for the laboratory diagnosis of multiple myeloma: A case study. J Clin Lab Anal. 2020 Sep;34(9):e23372. doi: 10.1002/jcla.23372. Epub 2020 Jun 16. PMID: 32548852; PMCID: PMC7521219.\u003c/li\u003e\n\u003cli\u003eLee SH, van der Weyden C, Mayson E, et al . Excessive EDTA induces morphologic changes in bone marrow smears that mimic specific features of dysplasia. Int J Lab Hematol. 2013 Apr;35(2):163-9. doi: 10.1111/ijlh.12015. Epub 2012 Oct 13.\u003c/li\u003e\n\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":true,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"scientific-reports","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"scirep","sideBox":"Learn more about [Scientific Reports](http://www.nature.com/srep/)","snPcode":"","submissionUrl":"","title":"Scientific Reports","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"Scientific Reports","inReviewEnabled":true,"inReviewRevisionsEnabled":true},"keywords":"","lastPublishedDoi":"10.21203/rs.3.rs-6195939/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-6195939/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eThe integration of artificial intelligence with bone marrow cytology represents a significant trend in the application of AI image recognition technology within the medical sector. Despite the current high accuracy of AI in cell identification, there remains a clinical need for fully automated AI cell recognition equipment that spans from sample processing to result generation. The source of the sample's origin as an influencing factor on the final results is a crucial consideration in equipment design. In this research, patient bone marrow fluid samples were processed into various types of anticoagulated bone marrow smears—EDTAK anticoagulated, sodium citrate anticoagulated, heparin lithium anticoagulated, and sodium citrate anticoagulated—and non-anticoagulated, and subsequently analyzed by AI devices. The findings revealed a significant lack of consistency in cell classification ratios and the total number of cells recognized between anticoagulated and non-anticoagulated bone marrow smear samples. This aspect must be taken into account when designing a fully automated AI-based bone marrow cell recognition device.\u003c/p\u003e","manuscriptTitle":"Consistency analysis of AI cell recognition results between non anticoagulant and anticoagulant marrow smear after Wright's staining","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-03-27 05:31:30","doi":"10.21203/rs.3.rs-6195939/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"decision","content":"Revision requested","date":"2025-12-10T17:26:32+00:00","index":"","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2025-11-27T10:40:53+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2025-11-23T21:58:40+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"262081603608545092010515377012238005580","date":"2025-11-21T16:26:50+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2025-11-19T12:19:58+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"95546780240339818499722448179427600674","date":"2025-11-18T10:18:40+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"252375092482287983181612731325183068879","date":"2025-11-18T09:06:45+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"173473230607820507462422773827327006656","date":"2025-11-18T09:03:07+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"176171675676382226799539440788848497243","date":"2025-11-18T08:27:43+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"181425676100279025440342568873462512688","date":"2025-03-25T06:48:52+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2025-03-24T20:52:16+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2025-03-24T20:34:25+00:00","index":"","fulltext":""},{"type":"editorInvited","content":"","date":"2025-03-21T19:26:55+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2025-03-21T00:47:47+00:00","index":"","fulltext":""},{"type":"submitted","content":"Scientific Reports","date":"2025-03-21T00:46:37+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"scientific-reports","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"scirep","sideBox":"Learn more about [Scientific Reports](http://www.nature.com/srep/)","snPcode":"","submissionUrl":"","title":"Scientific Reports","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"Scientific Reports","inReviewEnabled":true,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"bdc4e25b-ca60-403b-bdca-ce51be2c2ee7","owner":[],"postedDate":"March 27th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"under-review","subjectAreas":[{"id":46164701,"name":"Biological sciences/Computational biology and bioinformatics/Image processing"},{"id":46164702,"name":"Health sciences/Oncology/Cancer/Haematological cancer"}],"tags":[],"updatedAt":"2026-05-05T06:40:17+00:00","versionOfRecord":[],"versionCreatedAt":"2025-03-27 05:31:30","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-6195939","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-6195939","identity":"rs-6195939","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}