Computers in biology and medicine最新文献_第8页

Misidentifications in ayurvedic medicinal plants: Convolutional neural network (CNN) to overcome identification confusions 阿育吠陀药用植物的错误识别：卷积神经网络（CNN）克服识别混乱

IF 7 2区医学 Q1 BIOLOGY

Computers in biology and medicine

Pub Date : 2024-11-04 DOI: 10.1016/j.compbiomed.2024.109349

Nalaka Lankasena , Ruwani N. Nugara , Dhanesh Wisumperuma , Bathiya Seneviratne , Dilup Chandranimal , Kamal Perera

Plants are a vital ingredient of traditional medicine in Sri Lanka, and the quantity of medicinal plants used as a number differs in literature. Field identification of medicinal plants is carried out based on various plant characteristics. Conventional identification keys are available for plant identification, but it is a complex and time-consuming process that relies on manual observation, introducing inherent errors, particularly among individuals lacking extensive botanical expertise. This may cause uncertainty in the identification of plants due to lack of professional training, morphological similarity and nomenclatural confusion of plants. Such uncertainty may result in misidentifying another plant(s) as intended medicinal plants, which may lead to unsafe consequences. The objectives of the study were accomplished with the following three steps; listing the flowering plants used for medicinal purposes in Sri Lanka using multiple detailed botanical literatures, identifying medicinal plants that are confused with other medicinal or non-medicinal plants using literature and a questionnaire survey, and developing Convolutional Neural Networks (CNN) based technology to distinguish confusing plants. The study prepared a list of 1358 flowering plants cultivated and used in Sri Lanka as medicinal plants. Fifty-three medicinal plants that are confused with 63 medicinal and non-medicinal plant species were identified by two surveys. The CNN solution experimented with five species of the Bauhinia genus with close morphologically similar leaves with a high misidentification possibility. Using CNN EfficientNet-B0 Architecture for classification, four models were tested, and Model 4, which was trained using the augmented dataset with white-coloured background, resulted in a validation accuracy of 96.16%. The current study demonstrated the critical role of CNNs and EfficientNets in addressing misidentification issues in morphologically similar medicinal plants.

植物是斯里兰卡传统医药的重要成分，文献中使用的药用植物数量各不相同。药用植物的实地鉴定是根据植物的各种特征进行的。传统的识别钥匙可用于植物识别，但这是一个复杂而耗时的过程，依赖于人工观察，会产生固有误差，尤其是对缺乏广泛植物学专业知识的人而言。由于缺乏专业培训、植物形态相似和命名混乱等原因，这可能会导致植物识别的不确定性。这种不确定性可能会导致将其他植物误认为药用植物，从而导致不安全的后果。本研究的目标通过以下三个步骤实现：利用多种详细的植物学文献列出斯里兰卡的药用开花植物；利用文献和问卷调查确定与其他药用或非药用植物混淆的药用植物；开发基于卷积神经网络（CNN）的技术来区分混淆的植物。该研究编制了一份清单，列出了斯里兰卡栽培和用作药用植物的 1358 种开花植物。通过两次调查，确定了与 63 种药用和非药用植物混淆的 53 种药用植物。CNN 解决方案对紫荆属的五个物种进行了实验，这些物种的叶片形态十分相似，错误识别的可能性很高。使用 CNN EfficientNet-B0 架构进行分类，测试了四个模型，其中模型 4 是使用白色背景的增强数据集进行训练的，其验证准确率为 96.16%。本研究证明了 CNN 和 EfficientNets 在解决形态相似的药用植物的错误识别问题中的关键作用。

{"title":"Misidentifications in ayurvedic medicinal plants: Convolutional neural network (CNN) to overcome identification confusions","authors":"Nalaka Lankasena , Ruwani N. Nugara , Dhanesh Wisumperuma , Bathiya Seneviratne , Dilup Chandranimal , Kamal Perera","doi":"10.1016/j.compbiomed.2024.109349","DOIUrl":"10.1016/j.compbiomed.2024.109349","url":null,"abstract":"<div><div>Plants are a vital ingredient of traditional medicine in Sri Lanka, and the quantity of medicinal plants used as a number differs in literature. Field identification of medicinal plants is carried out based on various plant characteristics. Conventional identification keys are available for plant identification, but it is a complex and time-consuming process that relies on manual observation, introducing inherent errors, particularly among individuals lacking extensive botanical expertise. This may cause uncertainty in the identification of plants due to lack of professional training, morphological similarity and nomenclatural confusion of plants. Such uncertainty may result in misidentifying another plant(s) as intended medicinal plants, which may lead to unsafe consequences. The objectives of the study were accomplished with the following three steps; listing the flowering plants used for medicinal purposes in Sri Lanka using multiple detailed botanical literatures, identifying medicinal plants that are confused with other medicinal or non-medicinal plants using literature and a questionnaire survey, and developing Convolutional Neural Networks (CNN) based technology to distinguish confusing plants. The study prepared a list of 1358 flowering plants cultivated and used in Sri Lanka as medicinal plants. Fifty-three medicinal plants that are confused with 63 medicinal and non-medicinal plant species were identified by two surveys. The CNN solution experimented with five species of the <em>Bauhinia</em> genus with close morphologically similar leaves with a high misidentification possibility. Using CNN EfficientNet-B0 Architecture for classification, four models were tested, and Model 4, which was trained using the augmented dataset with white-coloured background, resulted in a validation accuracy of 96.16%. The current study demonstrated the critical role of CNNs and EfficientNets in addressing misidentification issues in morphologically similar medicinal plants.</div></div>","PeriodicalId":10578,"journal":{"name":"Computers in biology and medicine","volume":"183 ","pages":"Article 109349"},"PeriodicalIF":7.0,"publicationDate":"2024-11-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142579053","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

RADIANCE: Reliable and interpretable depression detection from speech using transformer RADIANCE：使用变压器从语音中进行可靠且可解释的抑郁检测。

IF 7 2区医学 Q1 BIOLOGY

Computers in biology and medicine

Pub Date : 2024-11-02 DOI: 10.1016/j.compbiomed.2024.109325

Anup Kumar Gupta, Ashutosh Dhamaniya, Puneet Gupta

Depression is a common but severe mental disorder that adversely impacts the ability of an individual to function normally in their day-to-day life. A majority of depressed individuals remain undiagnosed due to factors such as social stigma and a shortage of healthcare professionals. Consequently, several Machine Learning and Deep Learning (DL) models based on speech have been proposed for automatic depression detection, with the latter generally outperforming the former. However, DL models are blackbox and offer no transparency. In contrast, healthcare professionals prefer models that provide interpretability besides being accurate. In this direction, we propose a method RADIANCE (Reliable AnD InterpretAble depressioN deteCtion transformErs). RADIANCE incorporates a novel FilterBank VIsion Transformer (FBViT) network, which provides the symptoms of depression as interpretable features. Additionally, we employ a novel loss function that handles the class imbalance issue in the datasets. It also incorporates a penalty term that addresses the hierarchy of misclassification errors. We also propose a reliability predictor based on low-level descriptors that provides a reliability score to indicate the trustworthiness of the prediction by FBViT. Furthermore, in contrast to the conventional averaging and majority pooling, RADIANCE consolidates predictions from multiple clips of the input audio by intricately weighing each prediction based on its reliability score, ensuring a more accurate overall prediction. RADIANCE outperforms the state-of-the-art depression detection methods, achieving an accuracy of 89.36%, 80.36%, and 94.44% over the DAIC-WOZ, E-DAIC, and CMDC datasets, respectively. Further, RADIANCE achieves MAE scores of 3.27 and 5.04 on the DAIC-WOZ and E-DAIC datasets, respectively.

抑郁症是一种常见但严重的精神障碍，会对个人日常生活的正常能力产生不利影响。由于社会耻辱感和医疗保健专业人员短缺等因素，大多数抑郁症患者仍未得到诊断。因此，人们提出了几种基于语音的机器学习和深度学习（DL）模型，用于自动检测抑郁症，后者的表现普遍优于前者。然而，深度学习模型是黑盒子，不透明。相比之下，医疗专业人员更喜欢除了准确之外还能提供可解释性的模型。为此，我们提出了一种方法 RADIANCE（Reliable AnD InterpretAble DepressioN DeteCtion transformErs）。RADIANCE 融合了一个新颖的滤波库虚拟转换器（FBViT）网络，它提供了可解释的抑郁症状特征。此外，我们还采用了一种新颖的损失函数来处理数据集中的类不平衡问题。它还包含一个惩罚项，可解决误分类错误的层次问题。我们还提出了一种基于低级描述符的可靠性预测器，该预测器可提供可靠性评分，以显示 FBViT 预测的可信度。此外，与传统的平均法和多数池法不同，RADIANCE 通过基于可靠性得分对每个预测进行复杂的权衡，对来自多个输入音频片段的预测进行整合，从而确保整体预测更加准确。RADIANCE 优于最先进的抑郁检测方法，在 DAIC-WOZ、E-DAIC 和 CMDC 数据集上的准确率分别达到 89.36%、80.36% 和 94.44%。此外，RADIANCE 在 DAIC-WOZ 和 E-DAIC 数据集上的 MAE 分数分别为 3.27 和 5.04。

{"title":"RADIANCE: Reliable and interpretable depression detection from speech using transformer","authors":"Anup Kumar Gupta, Ashutosh Dhamaniya, Puneet Gupta","doi":"10.1016/j.compbiomed.2024.109325","DOIUrl":"10.1016/j.compbiomed.2024.109325","url":null,"abstract":"<div><div>Depression is a common but severe mental disorder that adversely impacts the ability of an individual to function normally in their day-to-day life. A majority of depressed individuals remain undiagnosed due to factors such as social stigma and a shortage of healthcare professionals. Consequently, several Machine Learning and Deep Learning (DL) models based on speech have been proposed for automatic depression detection, with the latter generally outperforming the former. However, DL models are blackbox and offer no transparency. In contrast, healthcare professionals prefer models that provide interpretability besides being accurate. In this direction, we propose a method <em>RADIANCE</em> (Reliable AnD InterpretAble depressioN deteCtion transformErs). <em>RADIANCE</em> incorporates a novel FilterBank VIsion Transformer (<em>FBViT</em>) network, which provides the symptoms of depression as interpretable features. Additionally, we employ a novel loss function that handles the class imbalance issue in the datasets. It also incorporates a penalty term that addresses the hierarchy of misclassification errors. We also propose a reliability predictor based on low-level descriptors that provides a reliability score to indicate the trustworthiness of the prediction by <em>FBViT</em>. Furthermore, in contrast to the conventional averaging and majority pooling, <em>RADIANCE</em> consolidates predictions from multiple clips of the input audio by intricately weighing each prediction based on its reliability score, ensuring a more accurate overall prediction. <em>RADIANCE</em> outperforms the state-of-the-art depression detection methods, achieving an accuracy of 89.36%, 80.36%, and 94.44% over the DAIC-WOZ, E-DAIC, and CMDC datasets, respectively. Further, <em>RADIANCE</em> achieves MAE scores of 3.27 and 5.04 on the DAIC-WOZ and E-DAIC datasets, respectively.</div></div>","PeriodicalId":10578,"journal":{"name":"Computers in biology and medicine","volume":"183 ","pages":"Article 109325"},"PeriodicalIF":7.0,"publicationDate":"2024-11-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142567731","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

FT-FEDTL: A fine-tuned feature-extracted deep transfer learning model for multi-class microwave-based brain tumor classification FT-FEDTL：用于基于微波的多类脑肿瘤分类的微调特征提取深度迁移学习模型。

IF 7 2区医学 Q1 BIOLOGY

Computers in biology and medicine

Pub Date : 2024-11-02 DOI: 10.1016/j.compbiomed.2024.109316

Amran Hossain , Rafiqul Islam , Mohammad Tariqul Islam , Phumin Kirawanich , Mohamed S. Soliman

The microwave brain imaging (MBI) system is an emerging technology used to detect brain tumors in their early stages. Multi-class microwave-based brain tumor (MBT) identification and classification are crucial due to the tumor's patterns and shape. Manual identification and categorization of the tumors from the images by physicians is a challenging task and consumes more time. Recently, to overcome these issues, the deep transfer learning (DTL) technique has been used to classify brain tumors efficiently. This paper proposes a Fine-tuned Feature Extracted Deep Transfer Learning Model called FT-FEDTL for multi-class MBT classification purposes. The main objective of this work is to suggest a better pathway for brain tumor diagnosis by designing an efficient DTL model that automatically identifies and categorizes the MBT images. The InceptionV3 architecture is utilized as a base for feature extraction in the proposed FT-FEDTL model. Thereafter, a fine-tuning method is applied to the additional five layers with hyperparameters. The fine-tuned layers are attached to the base model to enhance classification performance. The MBT data are collected from two sources and balanced by augmentation techniques to create a total of 4200 balanced datasets. Later, 80 % images are used for training, 20 % images are utilized for validation, and 80 samples of each class are used for testing the FT-FEDTL model for classifying tumors into six classes. We evaluated and compared the FT-FEDTL model with the three traditional non-CNN and seven pretrained models by applying an imbalanced and balanced dataset. The proposed model showed superior classification performance compared to other models for the balanced dataset. It attained an overall accuracy, recall, precision, specificity, and Fscore of 99.65 %, 99.16 %, 99.48 %, 99.10 %, and 99.23 %, respectively. The experimental outcomes ensure that the proposed model can be employed in biomedical applications to assist radiologists for multi-class MBT image classification purposes. The Anaconda distribution platform with Python 3.7 on the Windows 11 OS is used to implement the models.

微波脑成像（MBI）系统是一种用于早期检测脑肿瘤的新兴技术。由于肿瘤的形态和形状，基于微波的多类脑肿瘤（MBT）识别和分类至关重要。医生从图像中手动识别肿瘤并对其进行分类是一项具有挑战性的任务，而且会耗费更多时间。最近，为了克服这些问题，深度迁移学习（DTL）技术被用于对脑肿瘤进行有效分类。本文提出了一种名为 FT-FEDTL 的微调特征提取深度迁移学习模型，用于多类 MBT 分类。这项工作的主要目的是通过设计一种能自动识别和分类 MBT 图像的高效 DTL 模型，为脑肿瘤诊断提供更好的途径。在拟议的 FT-FEDTL 模型中，InceptionV3 架构被用作特征提取的基础。之后，对附加的五层采用超参数微调方法。微调后的层被附加到基础模型上，以提高分类性能。MBT 数据从两个来源收集，并通过增强技术进行平衡，共创建了 4200 个平衡数据集。之后，80% 的图像用于训练，20% 的图像用于验证，每类 80 个样本用于测试 FT-FEDTL 模型，以将肿瘤分为六类。我们采用不平衡的均衡数据集，对 FT-FEDTL 模型与三个传统非 CNN 模型和七个预训练模型进行了评估和比较。在平衡数据集上，与其他模型相比，所提出的模型显示出更优越的分类性能。它的总体准确率、召回率、精确度、特异性和 Fscore 分别达到了 99.65 %、99.16 %、99.48 %、99.10 % 和 99.23 %。实验结果确保了所提出的模型可以应用于生物医学领域，帮助放射科医生进行多类 MBT 图像分类。这些模型是在 Windows 11 操作系统上使用带有 Python 3.7 的 Anaconda 发布平台上实现的。

{"title":"FT-FEDTL: A fine-tuned feature-extracted deep transfer learning model for multi-class microwave-based brain tumor classification","authors":"Amran Hossain , Rafiqul Islam , Mohammad Tariqul Islam , Phumin Kirawanich , Mohamed S. Soliman","doi":"10.1016/j.compbiomed.2024.109316","DOIUrl":"10.1016/j.compbiomed.2024.109316","url":null,"abstract":"<div><div>The microwave brain imaging (MBI) system is an emerging technology used to detect brain tumors in their early stages. Multi-class microwave-based brain tumor (MBT) identification and classification are crucial due to the tumor's patterns and shape. Manual identification and categorization of the tumors from the images by physicians is a challenging task and consumes more time. Recently, to overcome these issues, the deep transfer learning (DTL) technique has been used to classify brain tumors efficiently. This paper proposes a Fine-tuned Feature Extracted Deep Transfer Learning Model called FT-FEDTL for multi-class MBT classification purposes. The main objective of this work is to suggest a better pathway for brain tumor diagnosis by designing an efficient DTL model that automatically identifies and categorizes the MBT images. The InceptionV3 architecture is utilized as a base for feature extraction in the proposed FT-FEDTL model. Thereafter, a fine-tuning method is applied to the additional five layers with hyperparameters. The fine-tuned layers are attached to the base model to enhance classification performance. The MBT data are collected from two sources and balanced by augmentation techniques to create a total of 4200 balanced datasets. Later, 80 % images are used for training, 20 % images are utilized for validation, and 80 samples of each class are used for testing the FT-FEDTL model for classifying tumors into six classes. We evaluated and compared the FT-FEDTL model with the three traditional non-CNN and seven pretrained models by applying an imbalanced and balanced dataset. The proposed model showed superior classification performance compared to other models for the balanced dataset. It attained an overall accuracy, recall, precision, specificity, and Fscore of 99.65 %, 99.16 %, 99.48 %, 99.10 %, and 99.23 %, respectively. The experimental outcomes ensure that the proposed model can be employed in biomedical applications to assist radiologists for multi-class MBT image classification purposes. The Anaconda distribution platform with Python 3.7 on the Windows 11 OS is used to implement the models.</div></div>","PeriodicalId":10578,"journal":{"name":"Computers in biology and medicine","volume":"183 ","pages":"Article 109316"},"PeriodicalIF":7.0,"publicationDate":"2024-11-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142567824","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Carotid single- and dual-layer stents reduce the wall adhesion of platelets by influencing flow and cellular transport 颈动脉单层和双层支架通过影响血流和细胞运输来减少壁上血小板的粘附。

IF 7 2区医学 Q1 BIOLOGY

Computers in biology and medicine

Pub Date : 2024-11-02 DOI: 10.1016/j.compbiomed.2024.109313

Christian J. Spieker , Axelle Y. Kern , Netanel Korin , Pierre H. Mangin , Alfons G. Hoekstra , Gábor Závodszky

An ongoing thrombosis on a ruptured atherosclerotic plaque in the carotid may cause stroke. The primary treatment for patients with tandem lesion is stenting. Dual-layer stents have been introduced as an alternative to single-layer stents for elective and emergent carotid artery stenting. While the dual-layer structure shows promise in reducing plaque prolapse through the stent struts and with it the occurrence of post-procedural embolism, there are early signs that this newer generation of stents is more thrombogenic. We investigate a single- and a dual-layer stent design to assess their influence on a set of thrombosis-related flow factors in a novel setup of combined experiments and simulations. The in vitro results reveal that both stents reduce thrombus formation by approximately 50% when human anticoagulated whole blood was perfused through macrofluidic flow chambers coated with either collagen or human atherosclerotic plaque homogenates. Simulations predict that the primary cause is reduced platelet presence in the vicinity of the wall, due to the influence of stents on flow and cellular transport. Both stents significantly alter the near-wall flow conditions, modifying shear rate, shear gradient, cell-free zones, and platelet availability. Additionally, the dual-layer stent has further increased local shear rates on the inner struts. It also displays increased stagnation zones and reduced recirculation between the outer-layer struts. Finally, the dual-layer stent shows further reduced adhesion over an atherosclerotic plaque coating. The novel approach presented here can be used to improve the design optimization process of cardiovascular stents in the future by allowing an in-depth study of the emerging flow characteristics and agonist transport.

颈动脉粥样硬化斑块破裂后持续形成的血栓可能导致中风。串联病变患者的主要治疗方法是支架植入术。双层支架作为单层支架的替代品已被引入择期和急诊颈动脉支架植入术。虽然双层结构有望减少斑块通过支架支柱的脱垂，从而减少术后栓塞的发生，但有早期迹象表明，这种新一代支架更容易形成血栓。我们研究了单层和双层支架的设计，通过实验和模拟相结合的新方法来评估它们对一系列血栓相关流动因素的影响。体外实验结果表明，当人体抗凝全血通过涂有胶原蛋白或人体动脉粥样硬化斑块匀浆的大流体流动室时，两种支架都能将血栓形成减少约 50%。模拟预测，主要原因是支架对血流和细胞运输的影响导致血管壁附近的血小板减少。两种支架都极大地改变了近壁流动条件，改变了剪切率、剪切梯度、无细胞区和血小板可用性。此外，双层支架进一步提高了内支架的局部剪切率。它还增加了停滞区，减少了外层支架之间的再循环。最后，双层支架在动脉粥样硬化斑块涂层上的粘附性进一步降低。通过深入研究新出现的流动特性和激动剂运输，本文介绍的新方法可用于改进未来心血管支架的优化设计过程。

{"title":"Carotid single- and dual-layer stents reduce the wall adhesion of platelets by influencing flow and cellular transport","authors":"Christian J. Spieker , Axelle Y. Kern , Netanel Korin , Pierre H. Mangin , Alfons G. Hoekstra , Gábor Závodszky","doi":"10.1016/j.compbiomed.2024.109313","DOIUrl":"10.1016/j.compbiomed.2024.109313","url":null,"abstract":"<div><div>An ongoing thrombosis on a ruptured atherosclerotic plaque in the carotid may cause stroke. The primary treatment for patients with tandem lesion is stenting. Dual-layer stents have been introduced as an alternative to single-layer stents for elective and emergent carotid artery stenting. While the dual-layer structure shows promise in reducing plaque prolapse through the stent struts and with it the occurrence of post-procedural embolism, there are early signs that this newer generation of stents is more thrombogenic. We investigate a single- and a dual-layer stent design to assess their influence on a set of thrombosis-related flow factors in a novel setup of combined experiments and simulations. The <em>in vitro</em> results reveal that both stents reduce thrombus formation by approximately 50% when human anticoagulated whole blood was perfused through macrofluidic flow chambers coated with either collagen or human atherosclerotic plaque homogenates. Simulations predict that the primary cause is reduced platelet presence in the vicinity of the wall, due to the influence of stents on flow and cellular transport. Both stents significantly alter the near-wall flow conditions, modifying shear rate, shear gradient, cell-free zones, and platelet availability. Additionally, the dual-layer stent has further increased local shear rates on the inner struts. It also displays increased stagnation zones and reduced recirculation between the outer-layer struts. Finally, the dual-layer stent shows further reduced adhesion over an atherosclerotic plaque coating. The novel approach presented here can be used to improve the design optimization process of cardiovascular stents in the future by allowing an in-depth study of the emerging flow characteristics and agonist transport.</div></div>","PeriodicalId":10578,"journal":{"name":"Computers in biology and medicine","volume":"183 ","pages":"Article 109313"},"PeriodicalIF":7.0,"publicationDate":"2024-11-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142567817","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Cost-effectiveness and cost-utility of community-based blinding fundus diseases screening with artificial intelligence: A modelling study from Shanghai, China 利用人工智能进行社区盲法眼底疾病筛查的成本效益和成本效用：中国上海的模型研究。

IF 7 2区医学 Q1 BIOLOGY

Computers in biology and medicine

Pub Date : 2024-11-02 DOI: 10.1016/j.compbiomed.2024.109329

Senlin Lin , Yingyan Ma , Liping Li , Yanwei Jiang , Yajun Peng , Tao Yu , Dan Qian , Yi Xu , Lina Lu , Yingyao Chen , Haidong Zou

Background

With application of artificial intelligence (AI) in the disease screening, process reengineering occurred simultaneously. Whether process reengineering deserves special emphasis in AI implementation in the community-based blinding fundus diseases screening is not clear.

Method

Cost-effectiveness and cost-utility analyses were performed employing decision-analytic Markov models. A hypothetical cohort of community residents was followed in the model over a period of 30 1-year Markov cycles, starting from the age of 60. The simulated cohort was based on work data of the Shanghai Digital Eye Disease Screening program (SDEDS). Three scenarios were compared: centralized screening with manual grading-based telemedicine systems (Scenario 1), centralized screening with an AI-assisted screening system (Scenario 2), and process reengineered screening with an AI-assisted screening system (Scenario 3). The main outcomes were incremental cost-effectiveness ratio (ICER) and incremental cost-utility ratio (ICUR).

Results

Compared with Scenario 1, Scenario 2 results in incremental 187.03 years of blindness avoided and incremental 106.78 QALYs at an additional cost of $ 490010.62 per 10,000 people screened, with an ICER of $2619.98 per year of blindness avoided and an ICUR of $4589.13 per QALY. Compared with Scenario 1, Scenario 3 results in incremental 187.03 years of blindness avoided and incremental 106.78 QALYs at an additional cost of $242313.23 per 10,000 people screened, with an ICER of $1295.60 per year of blindness avoided and an ICUR of $2269.35 per QALY. Although Scenario 2 and 3 could be considered cost-effective, the screening cost of Scenario 3 was 27.6 % and the total cost was 1.1 % lower, with the same expected effectiveness and utility. The probabilistic sensitivity analyses show that Scenario 3 dominated 69.1 % and 70.3 % of simulations under one and three times the local GDP per capita thresholds.

Conclusions

AI can improve the cost-effectiveness and cost-utility of screenings, especially when process reengineering is performed. Therefore, process reengineering is strongly recommended when AI is implemented.

背景：随着人工智能（AI）在疾病筛查中的应用，流程再造也同时发生。在社区盲法眼底疾病筛查中实施人工智能时，是否需要特别强调流程再造，目前尚不清楚：方法：采用决策分析马尔可夫模型进行成本效益和成本效用分析。模型中的社区居民假定队列从 60 岁开始，在 30 个 1 年马尔可夫周期内进行跟踪。模拟队列基于上海数字化眼病筛查项目（SDEDS）的工作数据。比较了三种方案：基于人工分级的远程医疗系统集中筛查（方案 1）、人工智能辅助筛查系统集中筛查（方案 2）和人工智能辅助筛查系统流程再造筛查（方案 3）。主要结果是增量成本效益比（ICER）和增量成本效用比（ICUR）：与方案 1 相比，方案 2 可增加 187.03 年的防盲时间，增加 106.78 QALY，每筛查 10000 人的额外成本为 490010.62 美元，每防盲一年的 ICER 为 2619.98 美元，每 QALY 的 ICUR 为 4589.13 美元。与方案 1 相比，方案 3 可增加 187.03 年的失明避免率和 106.78 QALY，每 1 万人筛查的额外成本为 242313.23 美元，每一年失明避免率的 ICER 为 1295.60 美元，每 QALY 的 ICUR 为 2269.35 美元。尽管方案 2 和方案 3 可被视为具有成本效益，但在预期效果和效用相同的情况下，方案 3 的筛查成本降低了 27.6%，总成本降低了 1.1%。概率敏感性分析表明，在当地人均 GDP 临界值为 1 倍和 3 倍的情况下，方案 3 分别占 69.1% 和 70.3% 的模拟比例：人工智能可以提高筛查的成本效益和成本效用，尤其是在进行流程再造时。因此，强烈建议在实施人工智能时进行流程再造。

{"title":"Cost-effectiveness and cost-utility of community-based blinding fundus diseases screening with artificial intelligence: A modelling study from Shanghai, China","authors":"Senlin Lin , Yingyan Ma , Liping Li , Yanwei Jiang , Yajun Peng , Tao Yu , Dan Qian , Yi Xu , Lina Lu , Yingyao Chen , Haidong Zou","doi":"10.1016/j.compbiomed.2024.109329","DOIUrl":"10.1016/j.compbiomed.2024.109329","url":null,"abstract":"<div><h3>Background</h3><div>With application of artificial intelligence (AI) in the disease screening, process reengineering occurred simultaneously. Whether process reengineering deserves special emphasis in AI implementation in the community-based blinding fundus diseases screening is not clear.</div></div><div><h3>Method</h3><div>Cost-effectiveness and cost-utility analyses were performed employing decision-analytic Markov models. A hypothetical cohort of community residents was followed in the model over a period of 30 1-year Markov cycles, starting from the age of 60. The simulated cohort was based on work data of the Shanghai Digital Eye Disease Screening program (SDEDS). Three scenarios were compared: centralized screening with manual grading-based telemedicine systems (Scenario 1), centralized screening with an AI-assisted screening system (Scenario 2), and process reengineered screening with an AI-assisted screening system (Scenario 3). The main outcomes were incremental cost-effectiveness ratio (ICER) and incremental cost-utility ratio (ICUR).</div></div><div><h3>Results</h3><div>Compared with Scenario 1, Scenario 2 results in incremental 187.03 years of blindness avoided and incremental 106.78 QALYs at an additional cost of $ 490010.62 per 10,000 people screened, with an ICER of $2619.98 per year of blindness avoided and an ICUR of $4589.13 per QALY. Compared with Scenario 1, Scenario 3 results in incremental 187.03 years of blindness avoided and incremental 106.78 QALYs at an additional cost of $242313.23 per 10,000 people screened, with an ICER of $1295.60 per year of blindness avoided and an ICUR of $2269.35 per QALY. Although Scenario 2 and 3 could be considered cost-effective, the screening cost of Scenario 3 was 27.6 % and the total cost was 1.1 % lower, with the same expected effectiveness and utility. The probabilistic sensitivity analyses show that Scenario 3 dominated 69.1 % and 70.3 % of simulations under one and three times the local GDP per capita thresholds.</div></div><div><h3>Conclusions</h3><div>AI can improve the cost-effectiveness and cost-utility of screenings, especially when process reengineering is performed. Therefore, process reengineering is strongly recommended when AI is implemented.</div></div>","PeriodicalId":10578,"journal":{"name":"Computers in biology and medicine","volume":"183 ","pages":"Article 109329"},"PeriodicalIF":7.0,"publicationDate":"2024-11-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142567823","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Multimodal brain tumor segmentation and classification from MRI scans based on optimized DeepLabV3+ and interpreted networks information fusion empowered with explainable AI. 基于可解释人工智能的优化 DeepLabV3+ 和解释网络信息融合的核磁共振成像扫描多模态脑肿瘤分割和分类。

IF 7 2区医学 Q1 BIOLOGY

Computers in biology and medicine

Pub Date : 2024-11-01 Epub Date: 2024-10-02 DOI: 10.1016/j.compbiomed.2024.109183

Muhammad Sami Ullah, Muhammad Attique Khan, Hussain Mobarak Albarakati, Robertas Damaševičius, Shrooq Alsenan

Explainable artificial intelligence (XAI) aims to offer machine learning (ML) methods that enable people to comprehend, properly trust, and create more explainable models. In medical imaging, XAI has been adopted to interpret deep learning black box models to demonstrate the trustworthiness of machine decisions and predictions. In this work, we proposed a deep learning and explainable AI-based framework for segmenting and classifying brain tumors. The proposed framework consists of two parts. The first part, encoder-decoder-based DeepLabv3+ architecture, is implemented with Bayesian Optimization (BO) based hyperparameter initialization. The different scales are performed, and features are extracted through the Atrous Spatial Pyramid Pooling (ASPP) technique. The extracted features are passed to the output layer for tumor segmentation. In the second part of the proposed framework, two customized models have been proposed named Inverted Residual Bottleneck 96 layers (IRB-96) and Inverted Residual Bottleneck Self-Attention (IRB-Self). Both models are trained on the selected brain tumor datasets and extracted features from the global average pooling and self-attention layers. Features are fused using a serial approach, and classification is performed. The BO-based hyperparameters optimization of the neural network classifiers is performed and the classification results have been optimized. An XAI method named LIME is implemented to check the interpretability of the proposed models. The experimental process of the proposed framework was performed on the Figshare dataset, and an average segmentation accuracy of 92.68 % and classification accuracy of 95.42 % were obtained, respectively. Compared with state-of-the-art techniques, the proposed framework shows improved accuracy.

可解释人工智能（XAI）旨在提供机器学习（ML）方法，使人们能够理解、正确信任并创建更多可解释的模型。在医学影像领域，XAI 已被用于解释深度学习黑盒模型，以证明机器决策和预测的可信度。在这项工作中，我们提出了一个基于深度学习和可解释人工智能的框架，用于分割和分类脑肿瘤。该框架由两部分组成。第一部分是基于 DeepLabv3+ 架构的编码器-解码器，通过基于贝叶斯优化（BO）的超参数初始化来实现。通过 Atrous Spatial Pyramid Pooling（ASPP）技术提取不同尺度的特征。提取的特征被传递到输出层进行肿瘤分割。在拟议框架的第二部分，提出了两个定制模型，分别名为 "96 层倒置残余瓶颈（IRB-96）"和 "倒置残余瓶颈自注意（IRB-Self）"。这两个模型都是在选定的脑肿瘤数据集上进行训练，并从全局平均汇集层和自我注意层提取特征。使用串行方法融合特征并进行分类。对神经网络分类器进行了基于 BO 的超参数优化，并对分类结果进行了优化。此外，还采用了一种名为 LIME 的 XAI 方法来检查所提模型的可解释性。在 Figshare 数据集上对所提出的框架进行了实验，结果显示平均分割准确率为 92.68%，平均分类准确率为 95.42%。与最先进的技术相比，所提出的框架提高了准确率。

{"title":"Multimodal brain tumor segmentation and classification from MRI scans based on optimized DeepLabV3+ and interpreted networks information fusion empowered with explainable AI.","authors":"Muhammad Sami Ullah, Muhammad Attique Khan, Hussain Mobarak Albarakati, Robertas Damaševičius, Shrooq Alsenan","doi":"10.1016/j.compbiomed.2024.109183","DOIUrl":"10.1016/j.compbiomed.2024.109183","url":null,"abstract":"<p><p>Explainable artificial intelligence (XAI) aims to offer machine learning (ML) methods that enable people to comprehend, properly trust, and create more explainable models. In medical imaging, XAI has been adopted to interpret deep learning black box models to demonstrate the trustworthiness of machine decisions and predictions. In this work, we proposed a deep learning and explainable AI-based framework for segmenting and classifying brain tumors. The proposed framework consists of two parts. The first part, encoder-decoder-based DeepLabv3+ architecture, is implemented with Bayesian Optimization (BO) based hyperparameter initialization. The different scales are performed, and features are extracted through the Atrous Spatial Pyramid Pooling (ASPP) technique. The extracted features are passed to the output layer for tumor segmentation. In the second part of the proposed framework, two customized models have been proposed named Inverted Residual Bottleneck 96 layers (IRB-96) and Inverted Residual Bottleneck Self-Attention (IRB-Self). Both models are trained on the selected brain tumor datasets and extracted features from the global average pooling and self-attention layers. Features are fused using a serial approach, and classification is performed. The BO-based hyperparameters optimization of the neural network classifiers is performed and the classification results have been optimized. An XAI method named LIME is implemented to check the interpretability of the proposed models. The experimental process of the proposed framework was performed on the Figshare dataset, and an average segmentation accuracy of 92.68 % and classification accuracy of 95.42 % were obtained, respectively. Compared with state-of-the-art techniques, the proposed framework shows improved accuracy.</p>","PeriodicalId":10578,"journal":{"name":"Computers in biology and medicine","volume":"182 ","pages":"109183"},"PeriodicalIF":7.0,"publicationDate":"2024-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142364647","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Corrigendum to "H2MaT-Unet:Hierarchical hybrid multi-axis transformer based Unet for medical image segmentation" [Comput. Biol. Med. 174C (2024) 108387]. H2MaT-Unet:Hierarchical hybrid multi-axis transformer based Unet for medical image segmentation" [Comput. Biol. Med. 174C (2024) 108387] 的更正。

IF 7 2区医学 Q1 BIOLOGY

Computers in biology and medicine

Pub Date : 2024-11-01 Epub Date: 2024-10-20 DOI: 10.1016/j.compbiomed.2024.109293

ZhiYong Ju, ZhongChen Zhou, ZiXiang Qi, Cheng Yi

引用次数: 0

Beyond the root: Geometric characterization for the diagnosis of syndromic heritable thoracic aortic diseases 超越根部：用于诊断综合遗传性胸主动脉疾病的几何特征。

IF 7 2区医学 Q1 BIOLOGY

Computers in biology and medicine

Pub Date : 2024-11-01 DOI: 10.1016/j.compbiomed.2024.109176

Pau Romero , Miguel Lozano , Lydia Dux-Santoy , Andrea Guala , Gisela Teixidó-Turà , Rafael Sebastián , Ignacio García-Fernández

Syndromic heritable thoracic aortic diseases (sHTAD), such as Marfan (MFS) or Loeys–Dietz (LDS) syndromes, involve high risk of life threatening aortic events. Diagnosis of syndromic features alone is difficult, and negative genetic tests do not necessarily exclude a genetic or hereditary condition. Periodic 3D imaging of the aorta is recommended in patients with aortic disease. Thus, an imaging-based approach aimed at identifying unique features of aortic geometry can be highly effective for diagnosing sHTAD and assessing risk. In this study, we present a method that can help identify the manifestations of sHTAD by focusing on the entire geometry of the thoracic aorta, rather than only using measurements of dilation of the aortic root. We analyze the geometric phenotype of 97 patients with genetically confirmed sHTAD (79 MF and 18 LDS) and of 45 healthy volunteers, using 3D aorta meshes obtained from phase contrast-enhanced magnetic resonance angiograms computed from 4D flow cardiac magnetic resonance. We build a geometric encoding of the aorta, based on a vessel coordinate system, and use several mathematical models to discriminate between controls and patients with sHTAD: a baseline scenario, based on aortic root dimensions only, a descriptor typically used in sHTAD patients; a low dimensional scenario, with a reduce encoding using principal component analysis; and a high-dimensional scenario, which included the full coefficient representation for geometry encoding, aiming to capture finer geometric details. The results indicate that considering the anatomy of the whole thoracic aorta can improve predictive ability. We achieve precision and sensitivity values over 0.8, with a specificity of over 70% in all the models used, while a single value classifiers (based only on aortic root diameter) demonstrated a trade-off between sensitivity and specificity. Using the mathematical properties of the vessel coordinate system representation, feature importance is mapped onto a set of anatomical traits that are used by the models to do the classification, thus providing interpretability of the results. This analysis indicates that in addition to the diameter of the aortic root, aortic elongation and a narrowing of the descending thoracic aorta may be markers of positive sHTAD.

综合征遗传性胸主动脉疾病（sHTAD），如马凡（MFS）或洛伊-迪茨（LDS）综合征，具有发生危及生命的主动脉事件的高风险。仅凭综合征特征进行诊断是很困难的，基因检测呈阴性并不一定能排除遗传或遗传性疾病。建议主动脉疾病患者定期进行主动脉三维成像检查。因此，旨在识别主动脉几何形状独特特征的成像方法对于诊断 sHTAD 和评估风险非常有效。在这项研究中，我们提出了一种方法，它可以通过关注整个胸主动脉的几何形状，而不仅仅是测量主动脉根部的扩张来帮助识别 sHTAD 的表现。我们利用从相位对比增强磁共振血管造影中获得的三维主动脉网格，分析了 97 名经基因证实的 sHTAD 患者（79 名中频患者和 18 名低频患者）和 45 名健康志愿者的几何表型。我们基于血管坐标系建立了主动脉的几何编码，并使用几种数学模型来区分对照组和 sHTAD 患者：基线方案，仅基于主动脉根部尺寸，这是通常用于 sHTAD 患者的描述符；低维方案，使用主成分分析进行还原编码；高维方案，包括几何编码的全系数表示，旨在捕捉更精细的几何细节。结果表明，考虑整个胸主动脉的解剖结构可以提高预测能力。我们使用的所有模型的精确度和灵敏度值都超过了 0.8，特异性超过了 70%，而单值分类器（仅基于主动脉根部直径）则在灵敏度和特异性之间进行了权衡。利用血管坐标系表示法的数学特性，特征的重要性被映射到一组解剖特征上，这些特征被模型用来进行分类，从而提供了结果的可解释性。这项分析表明，除了主动脉根部的直径外，主动脉伸长和降胸主动脉变窄也可能是 sHTAD 阳性的标志。

{"title":"Beyond the root: Geometric characterization for the diagnosis of syndromic heritable thoracic aortic diseases","authors":"Pau Romero , Miguel Lozano , Lydia Dux-Santoy , Andrea Guala , Gisela Teixidó-Turà , Rafael Sebastián , Ignacio García-Fernández","doi":"10.1016/j.compbiomed.2024.109176","DOIUrl":"10.1016/j.compbiomed.2024.109176","url":null,"abstract":"<div><div>Syndromic heritable thoracic aortic diseases (sHTAD), such as Marfan (MFS) or Loeys–Dietz (LDS) syndromes, involve high risk of life threatening aortic events. Diagnosis of syndromic features alone is difficult, and negative genetic tests do not necessarily exclude a genetic or hereditary condition. Periodic 3D imaging of the aorta is recommended in patients with aortic disease. Thus, an imaging-based approach aimed at identifying unique features of aortic geometry can be highly effective for diagnosing sHTAD and assessing risk. In this study, we present a method that can help identify the manifestations of sHTAD by focusing on the entire geometry of the thoracic aorta, rather than only using measurements of dilation of the aortic root. We analyze the geometric phenotype of 97 patients with genetically confirmed sHTAD (79 MF and 18 LDS) and of 45 healthy volunteers, using 3D aorta meshes obtained from phase contrast-enhanced magnetic resonance angiograms computed from 4D flow cardiac magnetic resonance. We build a geometric encoding of the aorta, based on a vessel coordinate system, and use several mathematical models to discriminate between controls and patients with sHTAD: a baseline scenario, based on aortic root dimensions only, a descriptor typically used in sHTAD patients; a low dimensional scenario, with a reduce encoding using principal component analysis; and a high-dimensional scenario, which included the full coefficient representation for geometry encoding, aiming to capture finer geometric details. The results indicate that considering the anatomy of the whole thoracic aorta can improve predictive ability. We achieve precision and sensitivity values over 0.8, with a specificity of over 70% in all the models used, while a single value classifiers (based only on aortic root diameter) demonstrated a trade-off between sensitivity and specificity. Using the mathematical properties of the vessel coordinate system representation, feature importance is mapped onto a set of anatomical traits that are used by the models to do the classification, thus providing interpretability of the results. This analysis indicates that in addition to the diameter of the aortic root, aortic elongation and a narrowing of the descending thoracic aorta may be markers of positive sHTAD.</div></div>","PeriodicalId":10578,"journal":{"name":"Computers in biology and medicine","volume":"182 ","pages":"Article 109176"},"PeriodicalIF":7.0,"publicationDate":"2024-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142616384","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Frontoparietal atrophy trajectories in cognitively unimpaired elderly individuals using longitudinal Bayesian clustering 利用纵向贝叶斯聚类研究认知功能未受损的老年人的额顶叶萎缩轨迹。

IF 7 2区医学 Q1 BIOLOGY

Computers in biology and medicine

Pub Date : 2024-11-01 DOI: 10.1016/j.compbiomed.2024.109190

G. Lorenzon , K. Poulakis , R. Mohanty , M. Kivipelto , M. Eriksdotter , D. Ferreira , E. Westman , for the Alzheimer's Disease Neuroimaging Initiative

Introduction

Frontal and/or parietal atrophy has been reported during aging. To disentangle the heterogeneity previously observed, this study aimed to uncover different clusters of grey matter profiles and trajectories within cognitively unimpaired individuals.

Methods

Structural magnetic resonance imaging (MRI) data of 307 Aβ-negative cognitively unimpaired individuals were modelled between ages 60–85 from three cohorts worldwide. We applied unsupervised clustering using a novel longitudinal Bayesian approach and characterized the clusters' cerebrovascular and cognitive profiles.

Results

Four clusters were identified with different grey matter profiles and atrophy trajectories. Differences were mainly observed in frontal and parietal brain regions. These distinct frontoparietal grey matter profiles and longitudinal trajectories were differently associated with cerebrovascular burden and cognitive decline.

Discussion

Our findings suggest a conciliation of the frontal and parietal theories of aging, uncovering coexisting frontoparietal GM patterns. This could have important future implications for better stratification and identification of at-risk individuals.

简介额叶和/或顶叶在衰老过程中出现萎缩。为了揭示之前观察到的异质性，本研究旨在发现认知功能未受损个体的灰质特征和轨迹的不同集群：对全球三个队列中年龄在 60-85 岁之间的 307 名 Aβ 阴性认知功能未受损者的结构性磁共振成像（MRI）数据进行建模。我们采用一种新颖的纵向贝叶斯方法进行了无监督聚类，并描述了这些聚类的脑血管和认知特征：结果：我们发现了四个具有不同灰质特征和萎缩轨迹的集群。主要在额叶和顶叶脑区观察到差异。这些不同的额顶叶灰质特征和纵向轨迹与脑血管负担和认知能力下降有着不同的关联：讨论：我们的研究结果表明，额叶和顶叶衰老理论是一致的，发现了共存的额顶灰质模式。这对未来更好地分层和识别高危人群具有重要意义。

{"title":"Frontoparietal atrophy trajectories in cognitively unimpaired elderly individuals using longitudinal Bayesian clustering","authors":"G. Lorenzon , K. Poulakis , R. Mohanty , M. Kivipelto , M. Eriksdotter , D. Ferreira , E. Westman , for the Alzheimer's Disease Neuroimaging Initiative","doi":"10.1016/j.compbiomed.2024.109190","DOIUrl":"10.1016/j.compbiomed.2024.109190","url":null,"abstract":"<div><h3>Introduction</h3><div>Frontal and/or parietal atrophy has been reported during aging. To disentangle the heterogeneity previously observed, this study aimed to uncover different clusters of grey matter profiles and trajectories within cognitively unimpaired individuals.</div></div><div><h3>Methods</h3><div>Structural magnetic resonance imaging (MRI) data of 307 Aβ-negative cognitively unimpaired individuals were modelled between ages 60–85 from three cohorts worldwide. We applied unsupervised clustering using a novel longitudinal Bayesian approach and characterized the clusters' cerebrovascular and cognitive profiles.</div></div><div><h3>Results</h3><div>Four clusters were identified with different grey matter profiles and atrophy trajectories. Differences were mainly observed in frontal and parietal brain regions. These distinct frontoparietal grey matter profiles and longitudinal trajectories were differently associated with cerebrovascular burden and cognitive decline.</div></div><div><h3>Discussion</h3><div>Our findings suggest a conciliation of the frontal and parietal theories of aging, uncovering coexisting frontoparietal GM patterns. This could have important future implications for better stratification and identification of at-risk individuals.</div></div>","PeriodicalId":10578,"journal":{"name":"Computers in biology and medicine","volume":"182 ","pages":"Article 109190"},"PeriodicalIF":7.0,"publicationDate":"2024-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142364732","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

On-site burn severity assessment using smartphone-captured color burn wound images. 使用智能手机捕捉的彩色烧伤创面图像进行现场烧伤严重程度评估。

IF 7 2区医学 Q1 BIOLOGY

Computers in biology and medicine

Pub Date : 2024-11-01 Epub Date: 2024-10-02 DOI: 10.1016/j.compbiomed.2024.109171

Xiayu Xu, Qilong Bu, Jingmeng Xie, Hang Li, Feng Xu, Jing Li

Accurate assessment of burn severity is crucial for the management of burn injuries. Currently, clinicians mainly rely on visual inspection to assess burns, characterized by notable inter-observer discrepancies. In this study, we introduce an innovative analysis platform using color burn wound images for automatic burn severity assessment. To do this, we propose a novel joint-task deep learning model, which is capable of simultaneously segmenting both burn regions and body parts, the two crucial components in calculating the percentage of total body surface area (%TBSA). Asymmetric attention mechanism is introduced, allowing attention guidance from the body part segmentation task to the burn region segmentation task. A user-friendly mobile application is developed to facilitate a fast assessment of burn severity at clinical settings. The proposed framework was evaluated on a dataset comprising 1340 color burn wound images captured on-site at clinical settings. The average Dice coefficients for burn depth segmentation and body part segmentation are 85.12 % and 85.36 %, respectively. The R² for %TBSA assessment is 0.9136. The source codes for the joint-task framework and the application are released on Github (https://github.com/xjtu-mia/BurnAnalysis). The proposed platform holds the potential to be widely used at clinical settings to facilitate a fast and precise burn assessment.

烧伤严重程度的准确评估对于烧伤的治疗至关重要。目前，临床医生主要依靠肉眼观察来评估烧伤，观察者之间存在明显差异。在本研究中，我们利用彩色烧伤创面图像引入了一个创新的分析平台，用于自动评估烧伤严重程度。为此，我们提出了一种新颖的联合任务深度学习模型，该模型能够同时分割烧伤区域和身体部位，这是计算总体表面积百分比（%TBSA）的两个关键部分。该模型引入了非对称注意力机制，可将注意力从身体部位分割任务引导到烧伤区域分割任务。为了便于在临床环境中快速评估烧伤严重程度，我们开发了一款用户友好型移动应用程序。所提出的框架在一个数据集上进行了评估，该数据集包括 1340 幅在临床环境中现场采集的彩色烧伤创面图像。烧伤深度分割和身体部位分割的平均 Dice 系数分别为 85.12 % 和 85.36 %。%TBSA评估的 R2 为 0.9136。联合任务框架和应用程序的源代码发布在 Github 上 (https://github.com/xjtu-mia/BurnAnalysis)。拟议的平台有望广泛应用于临床环境，以促进快速、精确的烧伤评估。

{"title":"On-site burn severity assessment using smartphone-captured color burn wound images.","authors":"Xiayu Xu, Qilong Bu, Jingmeng Xie, Hang Li, Feng Xu, Jing Li","doi":"10.1016/j.compbiomed.2024.109171","DOIUrl":"10.1016/j.compbiomed.2024.109171","url":null,"abstract":"<p><p>Accurate assessment of burn severity is crucial for the management of burn injuries. Currently, clinicians mainly rely on visual inspection to assess burns, characterized by notable inter-observer discrepancies. In this study, we introduce an innovative analysis platform using color burn wound images for automatic burn severity assessment. To do this, we propose a novel joint-task deep learning model, which is capable of simultaneously segmenting both burn regions and body parts, the two crucial components in calculating the percentage of total body surface area (%TBSA). Asymmetric attention mechanism is introduced, allowing attention guidance from the body part segmentation task to the burn region segmentation task. A user-friendly mobile application is developed to facilitate a fast assessment of burn severity at clinical settings. The proposed framework was evaluated on a dataset comprising 1340 color burn wound images captured on-site at clinical settings. The average Dice coefficients for burn depth segmentation and body part segmentation are 85.12 % and 85.36 %, respectively. The R<sup>2</sup> for %TBSA assessment is 0.9136. The source codes for the joint-task framework and the application are released on Github (https://github.com/xjtu-mia/BurnAnalysis). The proposed platform holds the potential to be widely used at clinical settings to facilitate a fast and precise burn assessment.</p>","PeriodicalId":10578,"journal":{"name":"Computers in biology and medicine","volume":"182 ","pages":"109171"},"PeriodicalIF":7.0,"publicationDate":"2024-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142371207","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0