Classifying motor imagery tasks via functional near-infrared spectroscopy (fNIRS) poses a significant challenge in brain-computer interface (BCI) research due to the high-dimensional nature of the signals. This study aimed to address this challenge by employing the common spatial pattern (CSP) algorithm to reduce input dimensions for support vector machine (SVM) and linear discriminant analysis (LDA) classifiers.
Methods
Data were collected from 15 healthy right-handed volunteers performing tasks involving left-hand motion, left-hand motor imagery, right-hand motion, and right-hand motor imagery. Signals from 20-channel fNIRS were utilized, with input features including statistical descriptors such as mean, variance, slope, skewness, and kurtosis. The CSP algorithm was integrated into both SVM and LDA classifiers to reduce dimensionality. The main statistical methods included classification accuracy assessment and comparison.
Results
Mean and slope were found to be the most discriminative features. Without CSP, SVM and LDA classifiers achieved average accuracies of 59.81 % ± 0.97 % and 69 % ± 11.42 %, respectively. However, with CSP integration, accuracies significantly improved to 81.63 % ± 0.99 % and 84.19 % ± 3.18 % for SVM and LDA, respectively. This value represents an increase of 21.82 % and 15.19 % in accuracy for SVM and LDA classifiers, respectively. Dimensionality reduction from 100 to 25 dimensions was achieved for SVM, leading to reduced computational complexity and faster calculation times. Additionally, the CSP technique enhanced LDA classifier accuracy by 3.31 % for both motion and motor imagery tasks.
Conclusion
Integration of the CSP algorithm may demonstrate promising potential for improving BCI systems' performance.
{"title":"Improvement of classification accuracy of functional near-infrared spectroscopy signals for hand motion and motor imagery using a common spatial pattern algorithm","authors":"Omid Asadi , Mahsan Hajihosseini , Sima Shirzadi , Zahra Einalou , Mehrdad Dadgostar","doi":"10.1016/j.imed.2024.05.004","DOIUrl":"10.1016/j.imed.2024.05.004","url":null,"abstract":"<div><h3>Objective</h3><div>Classifying motor imagery tasks via functional near-infrared spectroscopy (fNIRS) poses a significant challenge in brain-computer interface (BCI) research due to the high-dimensional nature of the signals. This study aimed to address this challenge by employing the common spatial pattern (CSP) algorithm to reduce input dimensions for support vector machine (SVM) and linear discriminant analysis (LDA) classifiers.</div></div><div><h3>Methods</h3><div>Data were collected from 15 healthy right-handed volunteers performing tasks involving left-hand motion, left-hand motor imagery, right-hand motion, and right-hand motor imagery. Signals from 20-channel fNIRS were utilized, with input features including statistical descriptors such as mean, variance, slope, skewness, and kurtosis. The CSP algorithm was integrated into both SVM and LDA classifiers to reduce dimensionality. The main statistical methods included classification accuracy assessment and comparison.</div></div><div><h3>Results</h3><div>Mean and slope were found to be the most discriminative features. Without CSP, SVM and LDA classifiers achieved average accuracies of 59.81 % ± 0.97 % and 69 % ± 11.42 %, respectively. However, with CSP integration, accuracies significantly improved to 81.63 % ± 0.99 % and 84.19 % ± 3.18 % for SVM and LDA, respectively. This value represents an increase of 21.82 % and 15.19 % in accuracy for SVM and LDA classifiers, respectively. Dimensionality reduction from 100 to 25 dimensions was achieved for SVM, leading to reduced computational complexity and faster calculation times. Additionally, the CSP technique enhanced LDA classifier accuracy by 3.31 % for both motion and motor imagery tasks.</div></div><div><h3>Conclusion</h3><div>Integration of the CSP algorithm may demonstrate promising potential for improving BCI systems' performance.</div></div>","PeriodicalId":73400,"journal":{"name":"Intelligent medicine","volume":"5 2","pages":"Pages 123-131"},"PeriodicalIF":4.4,"publicationDate":"2025-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144196280","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-05-01DOI: 10.1016/j.imed.2024.11.003
John Adeoye, Yuxiong Su
Background
Oral cancer is the most common head and neck malignancy and may develop from oral leukoplakia (OL) and oral lichenoid disease (OLD). Machine learning classifiers using structured (tabular) data have been employed to predict malignant transformation in OL and OLD. However, current models require improved discrimination, and their frameworks may limit feature fusion and multimodal risk prediction. Therefore, this study investigates whether tabular-to-image data conversion and deep learning (DL) based on convolutional neural networks (CNNs) can improve malignant transformation prediction compared to traditional classifiers.
Methods
This study used retrospective data of 1,010 patients with OL and OLD treated at Queen Mary Hospital, Hong Kong, from January 2003 to December 2023, to construct artificial intelligence-based models for oral cancer risk stratification in OL/OLD. Twenty-five input features and information on oral cancer development in OL/OLD were retrieved from electronic health records. Tabular-to-2D image data transformation was achieved by creating a feature matrix from encoded labels of the input variables arranged according to their correlation coefficient. Then, 2D images were used to populate five pre-trained DL models (VGG16, VGG19, MobileNetV2, ResNet50, and EfficientNet-B0). Area under the receiver operating characteristic curve (AUC), Brier scores, and net benefit of the DL models were calculated and compared to five traditional classifiers based on structured data and the binary epithelial dysplasia grading system (current method).
Results
This study found that the DL models had better AUC values (0.893–0.955) and Brier scores (0.072–0.106) compared to the traditional classifiers (AUC: 0.887–0.941 and Brier score: 0.074–0.136) during validation. During internal testing, VGG16 and VGG19 had better AUC values and Brier scores than other CNNs (AUC: 0.998–1.00; Brier score: 0.036–0.044) and the best traditional classifier (random forest) (AUC: 0.906; Brier score: 0.153). Additionally, VGG16 and VGG19 models outperformed random forest in discrimination and calibration during external testing (AUC: 1.00 vs. 0.976; Brier score: 0.022–0.034 vs. 0.129). The best CNNs also had better discriminatory performance and calibration than binary dysplasia grading at internal and external testing. Overall, decision curve analysis showed that the optimal DL models with transformed data had a higher net benefit than random forest and binary dysplasia grading.
Conclusion
Tabular-to-2D image data transformation may improve the use of structured input features for developing optimal intelligent models for oral cancer risk prediction in OL and OLD using convolutional networks. This approach may have the potential to robustly handle structured data in multimodal DL frameworks for oncological outcome prediction.
{"title":"Deep learning with data transformation improves cancer risk prediction in oral precancerous conditions","authors":"John Adeoye, Yuxiong Su","doi":"10.1016/j.imed.2024.11.003","DOIUrl":"10.1016/j.imed.2024.11.003","url":null,"abstract":"<div><h3>Background</h3><div>Oral cancer is the most common head and neck malignancy and may develop from oral leukoplakia (OL) and oral lichenoid disease (OLD). Machine learning classifiers using structured (tabular) data have been employed to predict malignant transformation in OL and OLD. However, current models require improved discrimination, and their frameworks may limit feature fusion and multimodal risk prediction. Therefore, this study investigates whether tabular-to-image data conversion and deep learning (DL) based on convolutional neural networks (CNNs) can improve malignant transformation prediction compared to traditional classifiers.</div></div><div><h3>Methods</h3><div>This study used retrospective data of 1,010 patients with OL and OLD treated at Queen Mary Hospital, Hong Kong, from January 2003 to December 2023, to construct artificial intelligence-based models for oral cancer risk stratification in OL/OLD. Twenty-five input features and information on oral cancer development in OL/OLD were retrieved from electronic health records. Tabular-to-2D image data transformation was achieved by creating a feature matrix from encoded labels of the input variables arranged according to their correlation coefficient. Then, 2D images were used to populate five pre-trained DL models (VGG16, VGG19, MobileNetV2, ResNet50, and EfficientNet-B0). Area under the receiver operating characteristic curve (AUC), Brier scores, and net benefit of the DL models were calculated and compared to five traditional classifiers based on structured data and the binary epithelial dysplasia grading system (current method).</div></div><div><h3>Results</h3><div>This study found that the DL models had better AUC values (0.893–0.955) and Brier scores (0.072–0.106) compared to the traditional classifiers (AUC: 0.887–0.941 and Brier score: 0.074–0.136) during validation. During internal testing, VGG16 and VGG19 had better AUC values and Brier scores than other CNNs (AUC: 0.998–1.00; Brier score: 0.036–0.044) and the best traditional classifier (random forest) (AUC: 0.906; Brier score: 0.153). Additionally, VGG16 and VGG19 models outperformed random forest in discrimination and calibration during external testing (AUC: 1.00 <em>vs</em>. 0.976; Brier score: 0.022–0.034 <em>vs</em>. 0.129). The best CNNs also had better discriminatory performance and calibration than binary dysplasia grading at internal and external testing. Overall, decision curve analysis showed that the optimal DL models with transformed data had a higher net benefit than random forest and binary dysplasia grading.</div></div><div><h3>Conclusion</h3><div>Tabular-to-2D image data transformation may improve the use of structured input features for developing optimal intelligent models for oral cancer risk prediction in OL and OLD using convolutional networks. This approach may have the potential to robustly handle structured data in multimodal DL frameworks for oncological outcome prediction.</div></div>","PeriodicalId":73400,"journal":{"name":"Intelligent medicine","volume":"5 2","pages":"Pages 141-150"},"PeriodicalIF":4.4,"publicationDate":"2025-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144196208","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-05-01DOI: 10.1016/j.imed.2025.02.001
Songwei Feng , Xia Yin , Yang Shen
The tumor microenvironment (TME) is a pivotal determinant of cancer progression and therapeutic response. The advent of individualized tumor therapy, based on the in-depth analysis of the TME, represents a revolutionary transformation in oncology. Artificial intelligence (AI) provides unparalleled capabilities to analyze and decipher the complexities of the TME through multi-omics integration, spatial modeling, and predictive analytics. By combining computational innovations with clinical insights, AI is driving a new paradigm in precision medicine. This editorial explores the transformative potential of AI in individualized tumor therapy, highlighting the groundbreaking applications and strategic directions to advance this field.
{"title":"Artificial intelligence-powered precision: Unveiling the tumor microenvironment for a new frontier in personalized cancer therapy","authors":"Songwei Feng , Xia Yin , Yang Shen","doi":"10.1016/j.imed.2025.02.001","DOIUrl":"10.1016/j.imed.2025.02.001","url":null,"abstract":"<div><div>The tumor microenvironment (TME) is a pivotal determinant of cancer progression and therapeutic response. The advent of individualized tumor therapy, based on the in-depth analysis of the TME, represents a revolutionary transformation in oncology. Artificial intelligence (AI) provides unparalleled capabilities to analyze and decipher the complexities of the TME through multi-omics integration, spatial modeling, and predictive analytics. By combining computational innovations with clinical insights, AI is driving a new paradigm in precision medicine. This editorial explores the transformative potential of AI in individualized tumor therapy, highlighting the groundbreaking applications and strategic directions to advance this field.</div></div>","PeriodicalId":73400,"journal":{"name":"Intelligent medicine","volume":"5 2","pages":"Pages 95-98"},"PeriodicalIF":4.4,"publicationDate":"2025-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144196282","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Osteosarcoma is a prevalent primary malignant bone tumor in children and adolescents, accounting for approximately 5 % of childhood malignancies. Because of its rarity and biological complexity, treatment breakthroughs for osteosarcoma have been limited. To advance research in this field, we aimed to construct the first comprehensive osteosarcoma knowledge graph (OSKG) using the PubMed database.
Methods
A systematic search of PubMed (2003–2023) using the keyword “osteosarcoma” yielded 25,415 abstracts. Leveraging BioBERT, pretrained on biomedical corpora and fine-tuned with osteosarcoma-specific manual annotations, we identified 16 entity types and 17 biological relationships. The extracted elements were synthesized to create the OSKG, resulting in a deep learning-based knowledge base to explore osteosarcoma pathogenesis and molecular mechanisms. We then developed a specialized question-answering system (knowledge graph question answering (KGQA)) powered by ChatGLM3. This system employs advanced natural language processing and incorporates the OSKG to ensure optimal response quality and accuracy.
Results
The pretrained BioBERT averaged > 92 % accuracy in entity and relationship training. Evaluation using 100 pairs of gold-standard quizzes showed that the final quiz system outperformed other large language models in accuracy and robustness.
Conclusion
The system is designed to provide accurate disease-related queries and answers, effectively facilitating knowledge acquisition and reasoning in medical research and clinical practice. This project offers a robust tool for osteosarcoma research and promotes the deep integration of knowledge graphs and artificial intelligence technologies in the medical field.
{"title":"Osteosarcoma knowledge graph question answering system: deep learning-based knowledge graph and large language model fusion","authors":"Lulu Zhang , Weisong Zhao , Zhiwei Cheng , Yafei Jiang , Kai Tian , Jia Shi , Zhenyu Jiang , Yingqi Hua","doi":"10.1016/j.imed.2024.12.001","DOIUrl":"10.1016/j.imed.2024.12.001","url":null,"abstract":"<div><h3>Objective</h3><div>Osteosarcoma is a prevalent primary malignant bone tumor in children and adolescents, accounting for approximately 5 % of childhood malignancies. Because of its rarity and biological complexity, treatment breakthroughs for osteosarcoma have been limited. To advance research in this field, we aimed to construct the first comprehensive osteosarcoma knowledge graph (OSKG) using the PubMed database.</div></div><div><h3>Methods</h3><div>A systematic search of PubMed (2003–2023) using the keyword “osteosarcoma” yielded 25,415 abstracts. Leveraging BioBERT, pretrained on biomedical corpora and fine-tuned with osteosarcoma-specific manual annotations, we identified 16 entity types and 17 biological relationships. The extracted elements were synthesized to create the OSKG, resulting in a deep learning-based knowledge base to explore osteosarcoma pathogenesis and molecular mechanisms. We then developed a specialized question-answering system (knowledge graph question answering (KGQA)) powered by ChatGLM3. This system employs advanced natural language processing and incorporates the OSKG to ensure optimal response quality and accuracy.</div></div><div><h3>Results</h3><div>The pretrained BioBERT averaged > 92 % accuracy in entity and relationship training. Evaluation using 100 pairs of gold-standard quizzes showed that the final quiz system outperformed other large language models in accuracy and robustness.</div></div><div><h3>Conclusion</h3><div>The system is designed to provide accurate disease-related queries and answers, effectively facilitating knowledge acquisition and reasoning in medical research and clinical practice. This project offers a robust tool for osteosarcoma research and promotes the deep integration of knowledge graphs and artificial intelligence technologies in the medical field.</div></div>","PeriodicalId":73400,"journal":{"name":"Intelligent medicine","volume":"5 2","pages":"Pages 99-110"},"PeriodicalIF":4.4,"publicationDate":"2025-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144196206","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-02-01DOI: 10.1016/j.imed.2024.09.007
Oluwafemi A. Adepoju , Ammar Usman Danazumi , Lamin BS Dibba , Bashiru Ibrahim , Salahuddin Iliyasu Gital , Joseph Gideon Ibrahim , Maliyogbinda L. Jibrailu , Emmanuel O. Balogun
Background
Monkeypox has become a significant public health concern owing to the recent epidemics and associated morbidity. The treatment is limited by the availability of drugs, especially in endemic communities. Computational methods can facilitate the discovery and development of new and effective therapies that are affordable. This study was aimed at identifying potential drug candidates from the SuperNatural chemical library against monkeypox virus essential proteins using computational methods.
Methods
We identified 7 highly conserved essential proteins involved in monkeypox virus (MPXV) replication, infectivity, and propagation as potential therapeutic targets. A library of 447 orally administrable drug-like compounds from the SuperNatural database was screened against the proteins for potential binders/ligands associations using virtual screening and molecular dynamics simulations.
Results
Our search identified hit compounds that mimicked the tecovirimat binding pose and outperformed it in binding affinity. Notably, resveratrol-3-O-D-glucopyranoside showed significant binding affinity to the viral protein F13L, a key protein involved in MPXV transmission. Extensive molecular dynamics simulations showed stable interactions between resveratrol-3-O-β-D-glucopyranoside and F13L, and other hit compounds with their respective targets.
Conclusion
Although the predicted interactions require further experimental validation, our results suggested that the identified compounds could be promising therapeutic candidates for the development of novel monkeypox drugs. These findings might underscore the significance of natural compounds in drug discovery and lay the foundation for developing novel antivirals against monkeypox.
背景：由于最近的流行和相关发病率，猴痘已成为一个重大的公共卫生问题。治疗受到药物供应的限制，特别是在流行社区。计算方法可以促进发现和开发新的、有效的、负担得起的治疗方法。本研究旨在利用计算方法从SuperNatural化学文库中鉴定抗猴痘病毒必需蛋白的潜在候选药物。方法鉴定7种高度保守的猴痘病毒（MPXV）复制、感染和繁殖必需蛋白，作为潜在的治疗靶点。利用虚拟筛选和分子动力学模拟，从SuperNatural数据库中筛选出447种口服给药类化合物，对潜在的结合物/配体结合蛋白进行筛选。结果我们的研究发现，hit化合物模仿了病毒素的结合姿势，并在结合亲和力上优于病毒素。值得注意的是，白藜芦醇-3- o - d -glucopyranoside与MPXV传播的关键蛋白F13L具有显著的结合亲和力。广泛的分子动力学模拟表明，白藜芦醇-3- o -β-D-glucopyranoside和F13L以及其他被击中的化合物与各自的靶标之间存在稳定的相互作用。结论虽然预测的相互作用需要进一步的实验验证，但我们的结果表明，鉴定的化合物可能是开发新型猴痘药物的有希望的治疗候选者。这些发现可能强调了天然化合物在药物发现中的重要性，并为开发新的抗猴痘抗病毒药物奠定了基础。
{"title":"Computational interrogation of natural compounds identified resveratrol-3-O-D-glucopyranoside as a potential inhibitor of essential monkeypox virus proteins","authors":"Oluwafemi A. Adepoju , Ammar Usman Danazumi , Lamin BS Dibba , Bashiru Ibrahim , Salahuddin Iliyasu Gital , Joseph Gideon Ibrahim , Maliyogbinda L. Jibrailu , Emmanuel O. Balogun","doi":"10.1016/j.imed.2024.09.007","DOIUrl":"10.1016/j.imed.2024.09.007","url":null,"abstract":"<div><h3>Background</h3><div>Monkeypox has become a significant public health concern owing to the recent epidemics and associated morbidity. The treatment is limited by the availability of drugs, especially in endemic communities. Computational methods can facilitate the discovery and development of new and effective therapies that are affordable. This study was aimed at identifying potential drug candidates from the SuperNatural chemical library against monkeypox virus essential proteins using computational methods.</div></div><div><h3>Methods</h3><div>We identified 7 highly conserved essential proteins involved in monkeypox virus (MPXV) replication, infectivity, and propagation as potential therapeutic targets. A library of 447 orally administrable drug-like compounds from the SuperNatural database was screened against the proteins for potential binders/ligands associations using virtual screening and molecular dynamics simulations.</div></div><div><h3>Results</h3><div>Our search identified hit compounds that mimicked the tecovirimat binding pose and outperformed it in binding affinity. Notably, resveratrol-3-O-D-glucopyranoside showed significant binding affinity to the viral protein F13L, a key protein involved in MPXV transmission. Extensive molecular dynamics simulations showed stable interactions between resveratrol-3-O-β-D-glucopyranoside and F13L, and other hit compounds with their respective targets.</div></div><div><h3>Conclusion</h3><div>Although the predicted interactions require further experimental validation, our results suggested that the identified compounds could be promising therapeutic candidates for the development of novel monkeypox drugs. These findings might underscore the significance of natural compounds in drug discovery and lay the foundation for developing novel antivirals against monkeypox.</div></div>","PeriodicalId":73400,"journal":{"name":"Intelligent medicine","volume":"5 1","pages":"Pages 5-13"},"PeriodicalIF":4.4,"publicationDate":"2025-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143454235","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-02-01DOI: 10.1016/j.imed.2024.11.001
Xinran Wang , Zhao Shi , Xiaoqian Ji , Bin Hu , Sui Chen , Longjiang Zhang
Objective
Intracranial aneurysm imaging artificial intelligence (AI) products have entered the clinical implementation phase, but the application status of them in Chinese hospitals remains unclear. A nationwide survey was conducted to explore the current status of intracranial aneurysm imaging AI products in hospitals across China.
Methods
Delphi method was used to develop a questionnaire, which was then distributed to the radiologists across China between September 3rd and 10th, 2023. Independent predictors of the adoption of these AI products, radiologists' attitudes, concerns and knowledge about these AI products were evaluated using logistic regression. Participants were categorized into seven groups based on Chinese geographical regions to compare the performance of these AI products in different geographical regions.
Results
After 3 rounds of Delphi discussion by 29 radiologists, the questionnaire was derived. A total of 961 radiologists from 777 different hospitals in 31 provinces across China completed the questionnaire. Among these hospitals, 45.4% (353/777) had introduced intracranial aneurysm imaging AI products. The most commonly reported concern with these AI products was poor specificity (265/446, 59.4%). The majority of respondents had basic (310/961, 42.0%) or intermediate (331/961, 44.9%) knowledge of AI products and they held positive attitudes (913/961, 95.0%) towards using them. Those who had received AI training were more likely to possess a higher level of knowledge about AI (odds ratio (OR) = 1.80, P = 0.04). For regional comparison, respondents in Central China and East China gave the highest ratings to the accuracy (OR = 2.41, P = 0.048 vs. OR=2.36, P = 0.02) and specificity (OR = 2.34, P = 0.046 vs. OR = 2.37, P = 0.02) of these AI products.
Conclusion
The intracranial aneurysm imaging AI products may be widely used in Chinese hospitals but vary by clinical scenarios and geographic position.
目的颅内动脉瘤成像人工智能（AI）产品已进入临床实施阶段，但在我国医院的应用状况尚不明朗。在全国范围内进行了一项调查，以了解中国各地医院颅内动脉瘤成像人工智能产品的现状。方法采用德尔菲法编制问卷，于2023年9月3日至10日向全国放射科医师发放。采用这些人工智能产品的独立预测因素，放射科医生对这些人工智能产品的态度、关注和知识，使用逻辑回归进行评估。参与者根据中国的地理区域分为七组，以比较这些人工智能产品在不同地理区域的性能。结果29名放射科医师经过3轮德尔菲讨论，得出问卷。来自中国31个省份777家不同医院的961名放射科医生完成了调查问卷。45.4%（353/777）的医院引进了颅内动脉瘤成像人工智能产品。这些人工智能产品最常见的问题是特异性差（265/446,59.4%）。大多数受访者对人工智能产品有基本（310/961,42.0%）或中级（331/961,44.9%）的知识，对使用人工智能产品持积极态度（913/961,95.0%）。接受过人工智能训练的人更有可能拥有更高水平的人工智能知识（比值比(OR) = 1.80, P = 0.04）。在区域比较中，华中和华东地区的受访者对这些人工智能产品的准确性（OR = 2.41, P = 0.048 vs. OR=2.36, P = 0.02）和特异性（OR = 2.34, P = 0.046 vs. OR= 2.37, P = 0.02）给出了最高的评分。结论颅内动脉瘤成像人工智能产品在我国医院可广泛应用，但因临床情况和地理位置而异。
{"title":"Nationwide survey of the status of artificial intelligence-based intracranial aneurysm detection systems","authors":"Xinran Wang , Zhao Shi , Xiaoqian Ji , Bin Hu , Sui Chen , Longjiang Zhang","doi":"10.1016/j.imed.2024.11.001","DOIUrl":"10.1016/j.imed.2024.11.001","url":null,"abstract":"<div><h3>Objective</h3><div>Intracranial aneurysm imaging artificial intelligence (AI) products have entered the clinical implementation phase, but the application status of them in Chinese hospitals remains unclear. A nationwide survey was conducted to explore the current status of intracranial aneurysm imaging AI products in hospitals across China.</div></div><div><h3>Methods</h3><div>Delphi method was used to develop a questionnaire, which was then distributed to the radiologists across China between September 3rd and 10th, 2023. Independent predictors of the adoption of these AI products, radiologists' attitudes, concerns and knowledge about these AI products were evaluated using logistic regression. Participants were categorized into seven groups based on Chinese geographical regions to compare the performance of these AI products in different geographical regions.</div></div><div><h3>Results</h3><div>After 3 rounds of Delphi discussion by 29 radiologists, the questionnaire was derived. A total of 961 radiologists from 777 different hospitals in 31 provinces across China completed the questionnaire. Among these hospitals, 45.4% (353/777) had introduced intracranial aneurysm imaging AI products. The most commonly reported concern with these AI products was poor specificity (265/446, 59.4%). The majority of respondents had basic (310/961, 42.0%) or intermediate (331/961, 44.9%) knowledge of AI products and they held positive attitudes (913/961, 95.0%) towards using them. Those who had received AI training were more likely to possess a higher level of knowledge about AI (odds ratio (OR) = 1.80, <em>P</em> = 0.04). For regional comparison, respondents in Central China and East China gave the highest ratings to the accuracy (OR = 2.41, <em>P</em> = 0.048 <em>vs</em>. OR=2.36, <em>P</em> = 0.02) and specificity (OR = 2.34, <em>P</em> = 0.046 <em>vs.</em> OR = 2.37, <em>P</em> = 0.02) of these AI products.</div></div><div><h3>Conclusion</h3><div>The intracranial aneurysm imaging AI products may be widely used in Chinese hospitals but vary by clinical scenarios and geographic position.</div></div>","PeriodicalId":73400,"journal":{"name":"Intelligent medicine","volume":"5 1","pages":"Pages 37-45"},"PeriodicalIF":4.4,"publicationDate":"2025-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143454241","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-02-01DOI: 10.1016/j.imed.2024.09.002
Xiaoping Chen , Feng Cao , Qing Wang , Zhewei Ye
With the rapid advancement of digitalization and intelligence in the medical field, a plethora of cutting-edge technologies are gradually being applied to revolutionize healthcare. In the medical data security privacy protection and artificial intelligence encryption computing, blockchain stands out due to its inherent characteristics of traceability, tamper-proofing, and high credibility. Although blockchain technology has been applied in various industries, its application in the medical field needs more driving force, and its development needs to be standardized. This clinical practice guideline is developed following the World Health Organization's recommended process, adopting Grading of Recommendations Assessment, Development and Evaluation in assessing evidence quality. Considering the integration of blockchain and medical scenarios, we focus on the value and implementability of practical medical applications and provide the guidance on the construction and application of medical blockchain. This practice guideline includesd 10 potential medical application scenarios and usage frameworks. It is worth highlighting that electronic medical record sharing, drug and device anti-counterfeiting, medical digital intellectual property protection, and public health management are considered to be the most easily implemented and effective medical scenarios. The recommendations in this guideline were formulated based on the consideration of stakeholder values and preferences, resource utilization, feasibility, and acceptability, may have a profound impact on the construction of medical blockchain-related scenarios in China and internationally.
Registration: Practice Guidance Registration for Transparency (PREPARE) website (http://www.guidelines-registry.cn) Registration No. PREPARE-2023CN637.
随着医疗领域数字化和智能化的快速发展，大量的尖端技术正在逐步应用于医疗保健革命。在医疗数据安全隐私保护和人工智能加密计算领域，区块链以其固有的可追溯性、防篡改性、高可信度等特点脱颖而出。虽然区块链技术已经应用于各个行业，但其在医疗领域的应用需要更多的驱动力，其发展需要规范化。本临床实践指南是根据世界卫生组织推荐的程序制定的，在评估证据质量时采用建议评估、发展和评价分级。考虑到区块链与医疗场景的融合，我们关注实际医疗应用的价值和可实施性，为医疗区块链的建设和应用提供指导。本实践指南包括10个潜在的医疗应用场景和使用框架。值得强调的是，电子病历共享、药品和器械防伪、医疗数字知识产权保护和公共卫生管理被认为是最容易实施和有效的医疗场景。本指南的建议基于利益相关者的价值观和偏好、资源利用、可行性和可接受性等方面的考虑而制定，可能对中国和国际医疗区块链相关场景的构建产生深远影响。注册：Practice Guidance Registration for Transparency （PREPARE）网站（http://www.guidelines-registry.cn）- 2023 cn637做好准备。
{"title":"2024 Chinese guideline on the construction and application of medical blockchain#","authors":"Xiaoping Chen , Feng Cao , Qing Wang , Zhewei Ye","doi":"10.1016/j.imed.2024.09.002","DOIUrl":"10.1016/j.imed.2024.09.002","url":null,"abstract":"<div><div>With the rapid advancement of digitalization and intelligence in the medical field, a plethora of cutting-edge technologies are gradually being applied to revolutionize healthcare. In the medical data security privacy protection and artificial intelligence encryption computing, blockchain stands out due to its inherent characteristics of traceability, tamper-proofing, and high credibility. Although blockchain technology has been applied in various industries, its application in the medical field needs more driving force, and its development needs to be standardized. This clinical practice guideline is developed following the World Health Organization's recommended process, adopting Grading of Recommendations Assessment, Development and Evaluation in assessing evidence quality. Considering the integration of blockchain and medical scenarios, we focus on the value and implementability of practical medical applications and provide the guidance on the construction and application of medical blockchain. This practice guideline includesd 10 potential medical application scenarios and usage frameworks. It is worth highlighting that electronic medical record sharing, drug and device anti-counterfeiting, medical digital intellectual property protection, and public health management are considered to be the most easily implemented and effective medical scenarios. The recommendations in this guideline were formulated based on the consideration of stakeholder values and preferences, resource utilization, feasibility, and acceptability, may have a profound impact on the construction of medical blockchain-related scenarios in China and internationally.</div><div><strong>Registration:</strong> Practice Guidance Registration for Transparency (PREPARE) website (<span><span>http://www.guidelines-registry.cn</span><svg><path></path></svg></span>) Registration No. PREPARE-2023CN637.</div></div>","PeriodicalId":73400,"journal":{"name":"Intelligent medicine","volume":"5 1","pages":"Pages 73-83"},"PeriodicalIF":4.4,"publicationDate":"2025-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143454287","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-02-01DOI: 10.1016/j.imed.2024.11.002
Jingjing Liu , Weijie Fan , Yi Yang , Qi Peng , Bingjun Ji , Luxing He , Yang Li , Jing Yuan , Wei Li , Xianqi Wang , Yi Wu , Chen Liu , Qingfang Gong , Mi He , Yeqin Fu , Dong Zhang , Si Zhang , Yongjian Nian
Background
Accurately identifying and localizing the five subtypes of intracranial hemorrhage (ICH) are crucial steps for subsequent clinical treatment; however, the lack of a large computed tomography (CT) dataset with annotations of the categorization and localization of ICH considerably limits the development of deep learning-based identification and localization methods. We aimed to construct this large dataset and develop a deep learning-based model to identify and localize the five ICH subtypes, including intraventricular hemorrhage (IVH), intraparenchymal hemorrhage (IPH), subdural hemorrhage (SDH), subarachnoid hemorrhage (SAH), and epidural hemorrhage (EDH), in non-contrast head CT scans.
Methods
Based on the public Radiological Society of North America (RSNA) 2019 dataset, we constructed a large CT dataset named RSNA 2019+ that was annotated for bleeding localization of the five ICH subtypes by three radiologists. An improved YOLOv8 architecture with the bidirectional feature pyramid network was proposed and trained using the RSNA 2019+ training dataset and evaluated on the RSNA 2019+ test dataset. The public CQ500, and two private datasets collected from the Xinqiao and Sunshine Union Hospitals, respectively, were also annotated to perform multicenter validation. Furthermore, the performance of the deep learning model was compared with that of four radiologists. Multiple performance metrics, including the average precision (AP), precision, recall and F1-score, were used for performance evaluation. The McNemar and chi-squared tests were performed, and the 95% Wilson confidence intervals were given for the precision and recall.
Results
There were 175,125; 4,707; 8,259; and 3,104 bounding boxes after annotation on the RSNA 2019+; CQ500+; and the PD 1 and PD 2 datasets, respectively. With an intersection-over-union threshold of 0.5, the APs of IVH, IPH, SAH, SDH and EDH are 0.852, 0.820, 0.574, 0.639, and 0.558, respectively, yielding a mean average precision (mAP) of 0.688 for our proposed deep learning model on the RSNA 2019+ test dataset. For the multicenter validation involving the three external datasets, the mAPs for CQ500, PD1, and PD2 were 0.594, 0.734, and 0.66, respectively, which is comparable to those of radiologist with eight years of experience in head CT interpretation.
Conclusion
The deep learning model developed from the constructed RSNA 2019+ dataset exhibited good potential in identifying and localizing the five ICH subtypes in CT slices and has the potential to assist in the clinical diagnosis.
{"title":"Deep learning-based identification and localization of intracranial hemorrhage in patients using a large annotated head computed tomography dataset: A retrospective multicenter study","authors":"Jingjing Liu , Weijie Fan , Yi Yang , Qi Peng , Bingjun Ji , Luxing He , Yang Li , Jing Yuan , Wei Li , Xianqi Wang , Yi Wu , Chen Liu , Qingfang Gong , Mi He , Yeqin Fu , Dong Zhang , Si Zhang , Yongjian Nian","doi":"10.1016/j.imed.2024.11.002","DOIUrl":"10.1016/j.imed.2024.11.002","url":null,"abstract":"<div><h3>Background</h3><div>Accurately identifying and localizing the five subtypes of intracranial hemorrhage (ICH) are crucial steps for subsequent clinical treatment; however, the lack of a large computed tomography (CT) dataset with annotations of the categorization and localization of ICH considerably limits the development of deep learning-based identification and localization methods. We aimed to construct this large dataset and develop a deep learning-based model to identify and localize the five ICH subtypes, including intraventricular hemorrhage (IVH), intraparenchymal hemorrhage (IPH), subdural hemorrhage (SDH), subarachnoid hemorrhage (SAH), and epidural hemorrhage (EDH), in non-contrast head CT scans.</div></div><div><h3>Methods</h3><div>Based on the public Radiological Society of North America (RSNA) 2019 dataset, we constructed a large CT dataset named RSNA 2019+ that was annotated for bleeding localization of the five ICH subtypes by three radiologists. An improved YOLOv8 architecture with the bidirectional feature pyramid network was proposed and trained using the RSNA 2019+ training dataset and evaluated on the RSNA 2019+ test dataset. The public CQ500, and two private datasets collected from the Xinqiao and Sunshine Union Hospitals, respectively, were also annotated to perform multicenter validation. Furthermore, the performance of the deep learning model was compared with that of four radiologists. Multiple performance metrics, including the average precision (AP), precision, recall and F1-score, were used for performance evaluation. The McNemar and chi-squared tests were performed, and the 95% Wilson confidence intervals were given for the precision and recall.</div></div><div><h3>Results</h3><div>There were 175,125; 4,707; 8,259; and 3,104 bounding boxes after annotation on the RSNA 2019+; CQ500+; and the PD 1 and PD 2 datasets, respectively. With an intersection-over-union threshold of 0.5, the APs of IVH, IPH, SAH, SDH and EDH are 0.852, 0.820, 0.574, 0.639, and 0.558, respectively, yielding a mean average precision (mAP) of 0.688 for our proposed deep learning model on the RSNA 2019+ test dataset. For the multicenter validation involving the three external datasets, the mAPs for CQ500, PD1, and PD2 were 0.594, 0.734, and 0.66, respectively, which is comparable to those of radiologist with eight years of experience in head CT interpretation.</div></div><div><h3>Conclusion</h3><div>The deep learning model developed from the constructed RSNA 2019+ dataset exhibited good potential in identifying and localizing the five ICH subtypes in CT slices and has the potential to assist in the clinical diagnosis.</div></div>","PeriodicalId":73400,"journal":{"name":"Intelligent medicine","volume":"5 1","pages":"Pages 14-22"},"PeriodicalIF":4.4,"publicationDate":"2025-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143454236","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-02-01DOI: 10.1016/j.imed.2024.09.004
Tamim M. Al-Hasan , Mohammad Noorizadeh , Faycal Bensaali , Nader Meskin , Ali Ait Hssain
Lung diseases pose a significant threat to public health worldwide, resulting in a substantial number of fatalities. Diseases such as chronic obstructive pulmonary disease and lung cancer constitute two of the three deadliest diseases worldwide, contributing to over 3 million deaths annually. This study offered a comparative analysis of different diagnostic techniques used for lung pathologies from an engineering standpoint. The review concentrated on intelligent detection methods, including electronic nose, computer vision (CV), or image processing, and biosensors such as graphene-field effect transistor (FET). The E-nose-based detection technique uses electronic sensors to recognize volatile organic compounds (VOCs) in the exhaled breath. These VOCs can aid in the diagnosis of lung pathologies such as pneumonia. The CV processing method involves the application of advanced imaging techniques and machine learning algorithms to scrutinize and diagnose lung pathologies and ventilator-associated pneumonia (VAP). Lastly, biosensors employ the exceptional properties of these materials to identify specific biomarkers in biological samples. This information can be used to diagnose lung pathologies and VAP. This study examined the current state-of-the-art methods and offers a comprehensive analysis of their advantages and disadvantages from an engineering perspective. The study underscored the potential of these techniques to enhance the diagnosis of lung pathologies and VAP and presents the advances in the field of smart biomedical applications. Additionally, it emphasized the necessity for further research to optimize their performance and clinical usefulness.
{"title":"Current trends and future orientation in diagnosing lung pathologies: A systematic survey","authors":"Tamim M. Al-Hasan , Mohammad Noorizadeh , Faycal Bensaali , Nader Meskin , Ali Ait Hssain","doi":"10.1016/j.imed.2024.09.004","DOIUrl":"10.1016/j.imed.2024.09.004","url":null,"abstract":"<div><div>Lung diseases pose a significant threat to public health worldwide, resulting in a substantial number of fatalities. Diseases such as chronic obstructive pulmonary disease and lung cancer constitute two of the three deadliest diseases worldwide, contributing to over 3 million deaths annually. This study offered a comparative analysis of different diagnostic techniques used for lung pathologies from an engineering standpoint. The review concentrated on intelligent detection methods, including electronic nose, computer vision (CV), or image processing, and biosensors such as graphene-field effect transistor (FET). The E-nose-based detection technique uses electronic sensors to recognize volatile organic compounds (VOCs) in the exhaled breath. These VOCs can aid in the diagnosis of lung pathologies such as pneumonia. The CV processing method involves the application of advanced imaging techniques and machine learning algorithms to scrutinize and diagnose lung pathologies and ventilator-associated pneumonia (VAP). Lastly, biosensors employ the exceptional properties of these materials to identify specific biomarkers in biological samples. This information can be used to diagnose lung pathologies and VAP. This study examined the current state-of-the-art methods and offers a comprehensive analysis of their advantages and disadvantages from an engineering perspective. The study underscored the potential of these techniques to enhance the diagnosis of lung pathologies and VAP and presents the advances in the field of smart biomedical applications. Additionally, it emphasized the necessity for further research to optimize their performance and clinical usefulness.</div></div>","PeriodicalId":73400,"journal":{"name":"Intelligent medicine","volume":"5 1","pages":"Pages 23-36"},"PeriodicalIF":4.4,"publicationDate":"2025-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143454237","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-02-01DOI: 10.1016/j.imed.2024.09.006
Longhui Li , Yifan Xiang , Xi Chen , Duoru Lin , Lanqin Zhao , Jun Xiao , Zhenzhe Lin , Jianyu Pang , Xiaotong Han , Lixue Liu , Yuxuan Wu , Zhenzhen Liu , Jingjing Chen , Jing Zhuang , Keming Yu , Haotian Lin
Background
Corneal stiffness abnormalities play an important role in the onset and progression of keratoconus. However, the limited availability of specialty devices for measuring corneal stiffness restricts their application in clinical practice. This study aimed to develop a machine learning (ML) model that can predict corneal stiffness based on ocular structures and investigate its efficacy in diagnosing keratoconus.
Methods
This retrospective study enrolled healthy individuals and keratoconus patients at the Zhongshan Ophthalmic Center from June 2018 to June 2021. Eleven features, including ocular structural parameters, intraocular pressure (IOP), and age were used to train ML regression models for predicting the stiffness parameter at first applanation (SP-A1) and the Corvis biomechanical index for Chinese populations (cCBI) measured by a Corvis ST device. Mean absolute errors (MAEs) and the area under the receiver operating characteristic curve (AUC) were used to evaluate the performance of the models. The diagnostic efficacy of the predicted SP-A1 and cCBI for keratoconus was evaluated by the AUC, net reclassification index (NRI), and integrated discrimination improvement (IDI).
Results
A total of 1,523 eyes were involved, of which 601 were diagnosed with keratoconus. The MAEs of the SP-A1 prediction were similar in cross-validation (8.95 mmHg/mm) and testing (10.65 mmHg/mm). The R2 value for the SP-A1 prediction exceeded 0.7, indicating that the performance was clinically acceptable. The AUC for the cCBI prediction was 0.935 (95% CI 0.906-0.963). The top three predictors for SP-A1 and cCBI were IOP, keratometry, and central corneal thickness. The addition of the predicted SP-A1 and cCBI significantly improved model performance in diagnosing keratoconus, with NRI of 0.607 (95% CI 0.367-0.812) and 0.188 (95% CI −0.022-0.398), and IDI of 0.028 (95% CI 0.006-0.048) and 0.045 (95% CI 0.018-0.072), respectively.
Conclusion
Our models predicted SP-A1 and cCBI relatively accurately in keratoconus and normal corneas. Moreover, the predicted SP-A1 and cCBI values significantly contributed to the diagnosis of keratoconus. These models could provide a potential alternative for evaluating corneal stiffness and thus facilitate keratoconus screening.
背景：角膜硬度异常在圆锥角膜的发生和发展中起重要作用。然而，用于测量角膜硬度的专用设备的有限可用性限制了它们在临床实践中的应用。本研究旨在建立一种基于眼部结构预测角膜硬度的机器学习（ML）模型，并探讨其在圆锥角膜诊断中的有效性。方法回顾性研究纳入2018年6月至2021年6月中山眼科中心的健康个体和圆锥角膜患者。使用眼结构参数、眼内压（IOP）和年龄等11个特征训练ML回归模型，预测首次压平时的刚度参数（SP-A1）和中国人群的Corvis生物力学指数（cCBI）。使用平均绝对误差（MAEs）和接收者工作特征曲线下面积（AUC）来评估模型的性能。通过AUC、净重分类指数（NRI）和综合判别改善（IDI）评价SP-A1和cCBI对圆锥角膜的诊断效果。结果共1523只眼受累，其中601只眼诊断为圆锥角膜。SP-A1预测的MAEs在交叉验证（8.95 mmHg/mm）和检验（10.65 mmHg/mm）中相似。SP-A1预测的R2值超过0.7，表明临床可接受。cCBI预测的AUC为0.935 （95% CI 0.906-0.963）。SP-A1和cCBI的前三个预测因子是IOP、角膜测量和角膜中央厚度。预测SP-A1和cCBI的加入显著提高了模型诊断圆锥角膜的性能，NRI分别为0.607 （95% CI 0.367-0.812）和0.188 (95% CI−0.022-0.398)，IDI分别为0.028 （95% CI 0.006-0.048）和0.045 （95% CI 0.018-0.072）。结论本模型对圆锥角膜和正常角膜SP-A1和cCBI的预测较为准确。此外，预测SP-A1和cCBI值对圆锥角膜的诊断有重要意义。这些模型可以为评估角膜硬度提供潜在的替代方法，从而促进圆锥角膜筛查。
{"title":"Machine learning model for predicting corneal stiffness and identifying keratoconus based on ocular structures","authors":"Longhui Li , Yifan Xiang , Xi Chen , Duoru Lin , Lanqin Zhao , Jun Xiao , Zhenzhe Lin , Jianyu Pang , Xiaotong Han , Lixue Liu , Yuxuan Wu , Zhenzhen Liu , Jingjing Chen , Jing Zhuang , Keming Yu , Haotian Lin","doi":"10.1016/j.imed.2024.09.006","DOIUrl":"10.1016/j.imed.2024.09.006","url":null,"abstract":"<div><h3>Background</h3><div>Corneal stiffness abnormalities play an important role in the onset and progression of keratoconus. However, the limited availability of specialty devices for measuring corneal stiffness restricts their application in clinical practice. This study aimed to develop a machine learning (ML) model that can predict corneal stiffness based on ocular structures and investigate its efficacy in diagnosing keratoconus.</div></div><div><h3>Methods</h3><div>This retrospective study enrolled healthy individuals and keratoconus patients at the Zhongshan Ophthalmic Center from June 2018 to June 2021. Eleven features, including ocular structural parameters, intraocular pressure (IOP), and age were used to train ML regression models for predicting the stiffness parameter at first applanation (SP-A1) and the Corvis biomechanical index for Chinese populations (cCBI) measured by a Corvis ST device. Mean absolute errors (MAEs) and the area under the receiver operating characteristic curve (AUC) were used to evaluate the performance of the models. The diagnostic efficacy of the predicted SP-A1 and cCBI for keratoconus was evaluated by the AUC, net reclassification index (NRI), and integrated discrimination improvement (IDI).</div></div><div><h3>Results</h3><div>A total of 1,523 eyes were involved, of which 601 were diagnosed with keratoconus. The MAEs of the SP-A1 prediction were similar in cross-validation (8.95 mmHg/mm) and testing (10.65 mmHg/mm). The R<sup>2</sup> value for the SP-A1 prediction exceeded 0.7, indicating that the performance was clinically acceptable. The AUC for the cCBI prediction was 0.935 (95% CI 0.906-0.963). The top three predictors for SP-A1 and cCBI were IOP, keratometry, and central corneal thickness. The addition of the predicted SP-A1 and cCBI significantly improved model performance in diagnosing keratoconus, with NRI of 0.607 (95% CI 0.367-0.812) and 0.188 (95% CI −0.022-0.398), and IDI of 0.028 (95% CI 0.006-0.048) and 0.045 (95% CI 0.018-0.072), respectively.</div></div><div><h3>Conclusion</h3><div>Our models predicted SP-A1 and cCBI relatively accurately in keratoconus and normal corneas. Moreover, the predicted SP-A1 and cCBI values significantly contributed to the diagnosis of keratoconus. These models could provide a potential alternative for evaluating corneal stiffness and thus facilitate keratoconus screening.</div></div>","PeriodicalId":73400,"journal":{"name":"Intelligent medicine","volume":"5 1","pages":"Pages 66-72"},"PeriodicalIF":4.4,"publicationDate":"2025-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143454245","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
扫码关注我们
求助内容：
应助结果提醒方式：
京公网安备 11010802042870号