Computer methods and programs in biomedicine最新文献

{"title":"Concept-based AI interpretability in physiological time-series data: Example of abnormality detection in electroencephalography","authors":"Alexander Brenner ,&nbsp;Felix Knispel ,&nbsp;Florian P. Fischer ,&nbsp;Peter Rossmanith ,&nbsp;Yvonne Weber ,&nbsp;Henner Koch ,&nbsp;Rainer Röhrig ,&nbsp;Julian Varghese ,&nbsp;Ekaterina Kutafina","doi":"10.1016/j.cmpb.2024.108448","DOIUrl":"10.1016/j.cmpb.2024.108448","url":null,"abstract":"<div><h3>Background and Objective</h3><div>Despite recent performance advancements, deep learning models are not yet adopted in clinical practice on a wide scale. The intrinsic intransparency of such systems is commonly cited as one major reason for this reluctance. This has motivated methods that aim to provide explanations of model functioning. Known limitations of feature-based explanations have led to an increased interest in concept-based interpretability. <em>Testing with Concept Activation Vectors</em> (TCAV) employs human-understandable, abstract concepts to explain model behavior. The method has previously been applied to the medical domain in the context of electronic health records, retinal fundus images and magnetic resonance imaging.</div></div><div><h3>Methods</h3><div>We explore the usage of TCAV for building interpretable models on physiological time series, using an example of abnormality detection in electroencephalography (EEG). For this purpose, we adopt the XceptionTime model, which is suitable for multi-channel physiological data of variable sizes. The model provides state-of-the-art performance on raw EEG data and is publically available. We propose and test several ideas regarding concept definition through metadata mining, using additional labeled EEG data and extracting interpretable signal characteristics in the form of frequencies. By including our own hospital data with analog labeling, we further evaluate the robustness of our approach.</div></div><div><h3>Results</h3><div>The tested concepts show a TCAV score distribution that is in line with the clinical expectations, i.e. concepts known to have strong links with EEG pathologies (such as epileptiform discharges) received higher scores than the neutral concepts (e.g. sex). The scores were consistent across the applied concept generation strategies.</div></div><div><h3>Conclusions</h3><div>TCAV has the potential to improve interpretability of deep learning applied to multi-channel signals as well as to detect possible biases in the data. Still, further work on developing the strategies for concept definition and validation on clinical physiological time series is needed to better understand how to extract clinically relevant information from the concept sensitivity scores.</div></div>","PeriodicalId":10624,"journal":{"name":"Computer methods and programs in biomedicine","volume":"257 ","pages":"Article 108448"},"PeriodicalIF":4.9,"publicationDate":"2024-09-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142420949","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}

{"title":"In silico data-based comparison of the accuracy and error source of various methods for noninvasively estimating central aortic blood pressure","authors":"Xujie Zhang ,&nbsp;Zhaojun Li ,&nbsp;Zhi Zhang ,&nbsp;Tianqi Wang ,&nbsp;Fuyou Liang","doi":"10.1016/j.cmpb.2024.108450","DOIUrl":"10.1016/j.cmpb.2024.108450","url":null,"abstract":"<div><h3>Background and objectives</h3><div>The higher clinical significance of central aortic blood pressure (CABP) compared to peripheral blood pressures has been extensively demonstrated. Accordingly, many methods for noninvasively estimating CABP have been proposed. However, there still lacks a systematic comparison of existing methods, especially in terms of how they differ in the ability to tolerate individual differences or measurement errors. The present study was designed to address this gap.</div></div><div><h3>Methods</h3><div>A large-scale ‘virtual subject’ dataset (n = 600) was created using a computational model of the cardiovascular system, and applied to examine several classical CABP estimation methods, including the direct method, generalized transfer function (GTF) method, n-point moving average (NPMA) method, second systolic pressure of periphery (SBP2) method, physical model-based wave analysis (MBWA) method, and suprasystolic cuff-based waveform reconstruction (SCWR) method. The errors of CABP estimation were analyzed and compared among methods with respect to the magnitude/distribution, correlations with physiological/hemodynamic factors, and sensitivities to noninvasive measurement errors.</div></div><div><h3>Results</h3><div>The errors of CABP estimation exhibited evident inter-method differences in terms of the mean and standard deviation (SD). Relatively, the estimation errors of the methods adopting pre-trained algorithms (i.e., the GTF and SCWR methods) were overall smaller and less sensitive to variations in physiological/hemodynamic conditions and random errors in noninvasive measurement of brachial arterial blood pressure (used for calibrating peripheral pulse wave). The performances of all the methods worsened following the introduction of random errors to peripheral pulse wave (used for deriving CABP), as characterized by the enlarged SD and/or increased mean of the estimation errors. Notably, the GTF and SCWR methods did not exhibit a better capability of tolerating pulse wave errors in comparison with other methods.</div></div><div><h3>Conclusions</h3><div>Classical noninvasive methods for estimating CABP were found to differ considerably in both the accuracy and error source, which provided theoretical evidence for understanding the specific advantages and disadvantages of each method. Knowledge about the method-specific error source and sensitivities of errors to different physiological/hemodynamic factors may contribute as theoretical references for interpreting clinical observations and exploring factors underlying large estimation errors, or provide guidance for optimizing existing methods or developing new methods.</div></div>","PeriodicalId":10624,"journal":{"name":"Computer methods and programs in biomedicine","volume":"257 ","pages":"Article 108450"},"PeriodicalIF":4.9,"publicationDate":"2024-09-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142380231","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}

{"title":"The impact of different degrees of stenosis on platelet deposition in the left anterior descending branch of the coronary artery","authors":"Yiming Zhao ,&nbsp;Haoyao Cao ,&nbsp;Yongtao Wei ,&nbsp;Tinghui Zheng","doi":"10.1016/j.cmpb.2024.108445","DOIUrl":"10.1016/j.cmpb.2024.108445","url":null,"abstract":"<div><h3>Background and objective</h3><div>This study aimed to investigate the impact of different stenotic degrees on platelet deposition in the left anterior descending branch of the coronary artery.</div></div><div><h3>Methods</h3><div>The idealized model of coronary artery stenosis of 30 %, 40 %, 50 %, 60 %, 70 % and four patient-specific models of 22.17 %, 34.88 %, 51.23 % and 62.96 % were established. A discrete phase model was used to calculate the deposition of platelet particles in blood.</div></div><div><h3>Results</h3><div>(1) As the stenotic degree increased from 30 % to 70 %, the maximum deposition rates were 4.23e^-² kg/(m² ·s), 3.47e^-² kg/(m² ·s), 0.14 kg/(m² ·s), 0.15 kg/(m² ·s), and 0.38 kg/(m² ·s), respectively. (2) The greater the stenotic degree, the more points of platelet deposition. (3) Platelets were mainly deposited at the proximal segment of mild stenosis. When the stenotic degree exceeded 50 %, the deposition position moved to the distal segment of the stenosis. (4) The results in the real coronary artery models were similar to those in the idealized model.</div></div><div><h3>Conclusion</h3><div>The study suggests that the location and number of platelet deposition are related to the degree of stenosis. Moderate to severe stenosis is more likely to spread downstream.</div></div>","PeriodicalId":10624,"journal":{"name":"Computer methods and programs in biomedicine","volume":"257 ","pages":"Article 108445"},"PeriodicalIF":4.9,"publicationDate":"2024-09-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142380232","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}

{"title":"Machine learning for post-liver transplant survival: Bridging the gap for long-term outcomes through temporal variation features","authors":"Kiruthika Balakrishnan ,&nbsp;Sawyer Olson ,&nbsp;Gyorgy Simon ,&nbsp;Lisiane Pruinelli","doi":"10.1016/j.cmpb.2024.108442","DOIUrl":"10.1016/j.cmpb.2024.108442","url":null,"abstract":"<div><h3>Background</h3><div>The long-term survival of liver transplant (LT) recipients is essential for optimizing organ allocation and estimating mortality outcomes. While models like the Model-for-End-Stage-Liver-Disease (MELD) predict 90-day mortality on the waiting list, they do not predict post-LT survival accurately. There is a need for predictive models that can forecast post-LT survival beyond the immediate period after transplantation.</div></div><div><h3>Method</h3><div>This study introduces new temporal variation features for predicting post-LT survival during the waiting list period. Cox Proportional-Hazards regression (CoxPH), Random Survival Forest (RSF), and Extreme Gradient Boosting (XGB) models are utilized, along with patient demographics and waiting list duration. Data from 716 LT patients from the University of Minnesota CTSI (2011–2021) are used to develop, evaluate, and compare post-LT survival prediction models.</div></div><div><h3>Results</h3><div>The temporal variation features, particularly when combined with the RSF model, proved most effective in predicting post-LT survival, with a C-index of 0.71 and an IBS of 0.151. This outperformed the predictive capability of the most recent MELD score, which had a C-index of &lt;0.51 in the same cohort.</div></div><div><h3>Conclusions</h3><div>Incorporating temporal variation features with the RSF model enhances long-term post-LT survival predictions. These insights can assist clinicians and patients in making more informed decisions about organ allocation and understanding the utility of LT, ultimately leading to improved patient outcomes.</div></div>","PeriodicalId":10624,"journal":{"name":"Computer methods and programs in biomedicine","volume":"257 ","pages":"Article 108442"},"PeriodicalIF":4.9,"publicationDate":"2024-09-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142379218","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}

{"title":"Pre-anesthetic brain network metrics as predictors of individual propofol sensitivity","authors":"Yun Zhang ,&nbsp;Fei Yan ,&nbsp;Qiang Wang ,&nbsp;Yubo Wang ,&nbsp;Liyu Huang","doi":"10.1016/j.cmpb.2024.108447","DOIUrl":"10.1016/j.cmpb.2024.108447","url":null,"abstract":"<div><h3>Background and Objective</h3><div>Numerous factors, including demographic characteristics, have been implicated in modulating individual sensitivity to propofol; however, substantial inter-individual differences persist even after accounting for these factors. This study thus aimed to explore whether pre-anesthesia brain functional network metrics correlate with an individual's sensitivity to propofol.</div></div><div><h3>Methods</h3><div>A total of 54 subjects, including 30 patients and 24 healthy volunteers, were enrolled. Propofol was administered via a target-controlled infusion device, and anesthesia depth was monitored using a bispectral index monitor. Sensitivity to propofol was quantified using the induction time, measured from infusion onset to the bispectral index, which reached 60. Brain functional network metrics indicative of functional integration and segregation, centrality, and network resilience were computed from pre-anesthetic 60-channel EEG recordings. Linear regression analysis and machine learning predictive models were applied to evaluate the contribution of pre-anesthesia network metrics in predicting individual sensitivity to propofol.</div></div><div><h3>Results</h3><div>Our analysis results revealed that subjects could be categorized into high- or low-sensitivity groups based on their induction time. Individuals with low sensitivity exhibited a greater network degree, clustering coefficient, global efficiency, and betweenness centrality, along with reduced modularity and assortativity coefficient in the alpha band. Furthermore, alpha band network metrics were significantly correlated with individual induction time. Leveraging these network metrics as features enabled the classification of individuals into high- or low-sensitivity groups with an accuracy of 94%.</div></div><div><h3>Conclusions</h3><div>Using a clinically relevant endpoint that signifies the level of anesthesia suitable for surgical procedures, this study underscored the robust correlation between pre-anesthesia alpha-band network metrics and individual sensitivity to propofol in a cohort that included both patients and healthy volunteers. Our findings offer preliminary insights into the potential utility of pre-anesthetic brain status assessment to predict propofol sensitivity on an individual basis, which may help to develop a more accurate personalized anesthesia plan.</div></div>","PeriodicalId":10624,"journal":{"name":"Computer methods and programs in biomedicine","volume":"257 ","pages":"Article 108447"},"PeriodicalIF":4.9,"publicationDate":"2024-09-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142375269","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}

{"title":"IG-Net: An Instrument-guided real-time semantic segmentation framework for prostate dissection during surgery for low rectal cancer","authors":"Bo Sun ,&nbsp;Zhen Sun ,&nbsp;Kexuan Li ,&nbsp;Xuehao Wang ,&nbsp;Guotao Wang ,&nbsp;Wenfeng Song ,&nbsp;Shuai Li ,&nbsp;Aimin Hao ,&nbsp;Yi Xiao","doi":"10.1016/j.cmpb.2024.108443","DOIUrl":"10.1016/j.cmpb.2024.108443","url":null,"abstract":"<div><h3>Background and Objective:</h3><div>Accurate prostate dissection is crucial in transanal surgery for patients with low rectal cancer. Improper dissection can lead to adverse events such as urethral injury, severely affecting the patient’s postoperative recovery. However, unclear boundaries, irregular shape of the prostate, and obstructive factors such as smoke present significant challenges for surgeons.</div></div><div><h3>Methods:</h3><div>Our innovative contribution lies in the introduction of a novel video semantic segmentation framework, IG-Net, which incorporates prior surgical instrument features for real-time and precise prostate segmentation. Specifically, we designed an instrument-guided module that calculates the surgeon’s region of attention based on instrument features, performs local segmentation, and integrates it with global segmentation to enhance performance. Additionally, we proposed a keyframe selection module that calculates the temporal correlations between consecutive frames based on instrument features. This module adaptively selects non-keyframe for feature fusion segmentation, reducing noise and optimizing speed.</div></div><div><h3>Results:</h3><div>To evaluate the performance of IG-Net, we constructed the most extensive dataset known to date, comprising 106 video clips and 6153 images. The experimental results reveal that this method achieves favorable performance, with 72.70% IoU, 82.02% Dice, and 35 FPS.</div></div><div><h3>Conclusions:</h3><div>For the task of prostate segmentation based on surgical videos, our proposed IG-Net surpasses all previous methods across multiple metrics. IG-Net balances segmentation accuracy and speed, demonstrating strong robustness against adverse factors.</div></div>","PeriodicalId":10624,"journal":{"name":"Computer methods and programs in biomedicine","volume":"257 ","pages":"Article 108443"},"PeriodicalIF":4.9,"publicationDate":"2024-09-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142379217","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}

{"title":"Classification of mindfulness experiences from gamma-band effective connectivity: Application of machine-learning algorithms on resting, breathing, and body scan","authors":"Ai-Ling Hsu ,&nbsp;Chun-Yu Wu ,&nbsp;Hei-Yin Hydra Ng ,&nbsp;Chun-Hsiang Chuang ,&nbsp;Chih-Mao Huang ,&nbsp;Changwei W. Wu ,&nbsp;Yi-Ping Chao","doi":"10.1016/j.cmpb.2024.108446","DOIUrl":"10.1016/j.cmpb.2024.108446","url":null,"abstract":"<div><h3>Background and Objective</h3><div>Practicing mindfulness is a mental process toward interoceptive awareness, achieving stress reduction and emotion regulation through brain-function alteration. Literature has shown that electroencephalography (EEG)-derived connectivity possesses the potential to differentiate brain functions between mindfulness naïve and mindfulness experienced, where such quantitative differentiation could benefit telediagnosis for mental health. However, there is no prior guidance in model selection targeting on the mindfulness-experience prediction. Here we hypothesized that the EEG effective connectivity could reach a good prediction performance in mindfulness experiences with brain interpretability.</div></div><div><h3>Methods</h3><div>We aimed at probing direct Directed Transfer Function (dDTF) to classify the participants’ history of mindfulness-based stress reduction (MBSR), and aimed at optimizing the prediction accuracy by comparing multiple machine learning (ML) algorithms. Targeting the gamma-band effective connectivity, we evaluated the EEG-based prediction of the mindfulness experiences across 7 machine learning (ML) algorithms and 3 sessions (i.e., resting, focus-breathing, and body-scan).</div></div><div><h3>Results</h3><div>The support vector machine and naïve Bayes classifiers exhibited significant accuracies above the chance level across all three sessions, and the decision tree algorithm reached the highest prediction accuracy of 91.7 % with the resting state, compared to the classification accuracies with the other two mindful states. We further conducted the analysis on essential EEG channels to preserve the classification accuracy, revealing that preserving just four channels (F7, F8, T7, and P7) out of 19 yielded the accuracy of 83.3 %. Delving into the contribution of connectivity features, specific connectivity features predominantly located in the frontal lobe contributed more to classifier construction, which aligned well with the existing mindfulness literature.</div></div><div><h3>Conclusion</h3><div>In the present study, we initiated a milestone of developing an EEG-based classifier to detect a person's mindfulness experience objectively. The prediction accuracy of the decision tree was optimal to differentiate the mindfulness experiences using the local resting-state EEG data. The suggested algorithm and key channels on the mindfulness-experience prediction may provide guidance for predicting mindfulness experiences using the EEG-based classification embedded in future wearable neurofeedback systems or plausible digital therapeutics.</div></div>","PeriodicalId":10624,"journal":{"name":"Computer methods and programs in biomedicine","volume":"257 ","pages":"Article 108446"},"PeriodicalIF":4.9,"publicationDate":"2024-09-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142380229","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}

{"title":"Unraveling the complex interplay between abnormal hemorheology and shape asymmetry in flow through stenotic arteries","authors":"Soumen Chakraborty ,&nbsp;Vishnu Teja Mantripragada ,&nbsp;Aranyak Chakravarty ,&nbsp;Debkalpa Goswami ,&nbsp;Antarip Poddar","doi":"10.1016/j.cmpb.2024.108437","DOIUrl":"10.1016/j.cmpb.2024.108437","url":null,"abstract":"<div><h3>Background and Objective:</h3><div>Stenosis or narrowing of arteries due to the buildup of plaque is a common occurrence in atherosclerosis and coronary artery disease (CAD), limiting blood flow to the heart and posing substantial cardiovascular risk. While the role of geometric irregularities in arterial stenosis is well-documented, the complex interplay between the abnormal hemorheology and asymmetric shape in flow characteristics remains unexplored.</div></div><div><h3>Methods:</h3><div>This study investigates the influence of varying hematocrit (Hct) levels, often caused by conditions such as diabetes and anemia, on flow patterns in an idealized eccentric stenotic artery using computational fluid dynamics simulations. We consider three physiological levels of Hct, 25%, 45%, and 65%, representing anemia, healthy, and diabetic conditions, respectively. The numerical simulations are performed for different combinations of shape eccentricity and blood rheological parameters, and hemodynamic indicators such as wall shear stress (WSS), oscillatory shear index (OSI), are relative residence time (RRT) are calculated to assess the arterial health.</div></div><div><h3>Results:</h3><div>Our results reveal the significant influence of <span><math><mrow><mi>H</mi><mi>c</mi><mi>t</mi></mrow></math></span> level on stenosis progression. CAD patients with anemia are exposed to lower WSS and higher OSI, which may increase the propensity for plaque progression and rupture. However, for CAD patients with high <span><math><mrow><mi>H</mi><mi>c</mi><mi>t</mi></mrow></math></span> level — as is often the case in diabetes — the WSS at the minimal lumen area increases rapidly, which may also lead to plaque rupture and cause adverse events such as heart attacks. These disturbances promote endothelial dysfunction, inflammation, and thrombus formation, thereby intensifying cardiovascular risk.</div></div><div><h3>Conclusions:</h3><div>Our findings underscore the significance of incorporating hemorheological parameters, such as <span><math><mrow><mi>H</mi><mi>c</mi><mi>t</mi></mrow></math></span>, into computational models for accurate assessment of flow dynamics. We envision that insights gained from this study will inform the development of tailored treatment strategies and interventions in CAD patients with common comorbidities such as diabetes and anemia, thus mitigating the adverse effects of abnormal hemorheology and reducing the ever-growing burden of cardiovascular diseases.</div></div>","PeriodicalId":10624,"journal":{"name":"Computer methods and programs in biomedicine","volume":"257 ","pages":"Article 108437"},"PeriodicalIF":4.9,"publicationDate":"2024-09-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142364651","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}

{"title":"Paying attention to uncertainty: A stochastic multimodal transformers for post-traumatic stress disorder detection using video","authors":"Mamadou Dia,&nbsp;Ghazaleh Khodabandelou,&nbsp;Alice Othmani","doi":"10.1016/j.cmpb.2024.108439","DOIUrl":"10.1016/j.cmpb.2024.108439","url":null,"abstract":"<div><h3>Background and Objectives:</h3><div>Post-traumatic stress disorder is a debilitating psychological condition that can manifest following exposure to traumatic events. It affects individuals from diverse backgrounds and is associated with various symptoms, including intrusive thoughts, nightmares, hyperarousal, and avoidance behaviors.</div></div><div><h3>Methods:</h3><div>To address this challenge this study proposes a decision support system powered by a novel multimodal deep learning approach, based on a stochastic Transformer and video data. This Transformer has the ability to take advantage of its stochastic activation function and layers that allow it to learn sparse representations of the inputs. The method leverages a combination of low-level features extracted using three modalities, including Mel-frequency cepstral coefficients extracted from audio recordings, Facial Action Units captured from facial expressions, and textual data obtained from the audio transcription. By considering these modalities, our proposed model captures a comprehensive range of information related to post-traumatic stress disorder symptoms, including vocal cues, facial expressions, and linguistic content.</div></div><div><h3>Results:</h3><div>The deep learning model was trained and evaluated on the eDAIC dataset, which consists of clinical interviews with individuals with and without post-traumatic disorder. The model achieved state-of-the-art results, demonstrating its effectiveness in accurately detecting PTSD, showing an impressive Root Mean Square Error of 1.98, and a Concordance Correlation Coefficient of 0.722, signifying the model’s superior performance compared to existing approaches.</div></div><div><h3>Conclusion:</h3><div>This work introduces a new method for post-traumatic stress disorder detection from videos by utilizing a multimodal stochastic Transformer model. The model makes use of a variety of modalities, such as text, audio, and visual data, to gather comprehensive and varied information in order to make the detection.</div></div>","PeriodicalId":10624,"journal":{"name":"Computer methods and programs in biomedicine","volume":"257 ","pages":"Article 108439"},"PeriodicalIF":4.9,"publicationDate":"2024-09-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142326265","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}

{"title":"GLCV-NET: An automatic diagnosis system for advanced liver fibrosis using global–local cross view in B-mode ultrasound images","authors":"Bianzhe Wu ,&nbsp;ZeRong Huang ,&nbsp;Jinglin Liang ,&nbsp;Hong Yang ,&nbsp;Wei Wang ,&nbsp;Shuangping Huang ,&nbsp;LiDa Chen ,&nbsp;Qinghua Huang","doi":"10.1016/j.cmpb.2024.108440","DOIUrl":"10.1016/j.cmpb.2024.108440","url":null,"abstract":"<div><h3>Background and objective</h3><div>: Advanced liver fibrosis is a critical stage in the evaluation of chronic liver disease (CLD), holding clinical significance in the development of treatment strategies and estimating the disease progression.</div></div><div><h3>Methods:</h3><div>This paper proposes an innovative Global–Local Cross-View Network (GLCV-Net) for the automatic diagnosis of advanced liver fibrosis from ultrasound (US) B-mode images. The proposed method consists of three main components: 1. A Segmentation-enhanced Global Hybrid Feature Extractor for segmenting the liver parenchyma and extracting global features; 2. A Heatmap-weighted Local Feature Extractor for selecting candidate regions and automatically identifying suspicious areas to construct local features; 3. A Scale-adaptive Fusion Module to balance the contributions of global and local scales in evaluating advanced liver fibrosis.</div></div><div><h3>Results:</h3><div>The predictive performance of the model was validated on an internal dataset of 1003 chronic liver disease (CLD) patients and an external dataset of 46 CLD patients, both subjected to liver fibrosis staging through pathological assessment. On the internal dataset, GLCV-Net achieved 86.9% accuracy, 85.0% recall, 85.4% precision, and 85.2% F1-score. Further validation on the external dataset confirmed its robustness, with scores of 86.1% in accuracy, 83.1% in recall, 80.8% in precision, and 81.9% in F1-score.</div></div><div><h3>Conclusion:</h3><div>These results underscore the GLCV-Net’s potential as a promising approach for non-invasively and accurately diagnosing advanced liver fibrosis in CLD patients, breaking through the limitations of traditional methods by integrating global and local information of liver fibrosis, significantly enhancing diagnostic accuracy.</div></div>","PeriodicalId":10624,"journal":{"name":"Computer methods and programs in biomedicine","volume":"257 ","pages":"Article 108440"},"PeriodicalIF":4.9,"publicationDate":"2024-09-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142388709","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}