Journal of medical artificial intelligence最新文献

{"title":"Efficient labelling for efficient deep learning: the benefit of a multiple-image-ranking method to generate high volume training data applied to ventricular slice level classification in cardiac MRI.","authors":"Sameer Zaman, Kavitha Vimalesvaran, James P Howard, Digby Chappell, Marta Varela, Nicholas S Peters, Darrel P Francis, Anil A Bharath, Nick W F Linton, Graham D Cole","doi":"10.21037/jmai-22-55","DOIUrl":"10.21037/jmai-22-55","url":null,"abstract":"<p><strong>Background: </strong>Getting the most value from expert clinicians' limited labelling time is a major challenge for artificial intelligence (AI) development in clinical imaging. We present a novel method for ground-truth labelling of cardiac magnetic resonance imaging (CMR) image data by leveraging multiple clinician experts ranking multiple images on a single ordinal axis, rather than manual labelling of one image at a time. We apply this strategy to train a deep learning (DL) model to classify the anatomical position of CMR images. This allows the automated removal of slices that do not contain the left ventricular (LV) myocardium.</p><p><strong>Methods: </strong>Anonymised LV short-axis slices from 300 random scans (3,552 individual images) were extracted. Each image's anatomical position relative to the LV was labelled using two different strategies performed for 5 hours each: (I) 'one-image-at-a-time': each image labelled according to its position: 'too basal', 'LV', or 'too apical' individually by one of three experts; and (II) 'multiple-image-ranking': three independent experts ordered slices according to their relative position from 'most-basal' to 'most apical' in batches of eight until each image had been viewed at least 3 times. Two convolutional neural networks were trained for a three-way classification task (each model using data from one labelling strategy). The models' performance was evaluated by accuracy, F1-score, and area under the receiver operating characteristics curve (ROC AUC).</p><p><strong>Results: </strong>After excluding images with artefact, 3,323 images were labelled by both strategies. The model trained using labels from the 'multiple-image-ranking strategy' performed better than the model using the 'one-image-at-a-time' labelling strategy (accuracy 86% <i>vs.</i> 72%, P=0.02; F1-score 0.86 <i>vs.</i> 0.75; ROC AUC 0.95 <i>vs.</i> 0.86). For expert clinicians performing this task manually the intra-observer variability was low (Cohen's κ=0.90), but the inter-observer variability was higher (Cohen's κ=0.77).</p><p><strong>Conclusions: </strong>We present proof of concept that, given the same clinician labelling effort, comparing multiple images side-by-side using a 'multiple-image-ranking' strategy achieves ground truth labels for DL more accurately than by classifying images individually. We demonstrate a potential clinical application: the automatic removal of unrequired CMR images. This leads to increased efficiency by focussing human and machine attention on images which are needed to answer clinical questions.</p>","PeriodicalId":73815,"journal":{"name":"Journal of medical artificial intelligence","volume":"6 ","pages":"4"},"PeriodicalIF":0.0,"publicationDate":"2023-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7614685/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"9710776","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}

{"title":"Defining functional requirements for a patient-centric computerized glaucoma treatment and care ecosystem","authors":"N. Goldmann, S. Skalicky, R. Weinreb, R. P. Paletta Guedes, C. Baudouin, Xiulan Zhang, Aukje van Gestel, E. Blumenthal, P. Kaufman, R. Rothman, Ana Maria Vasquez, P. Harasymowycz, D. Welsbie, I. Goldberg","doi":"10.21037/jmai-22-33","DOIUrl":"https://doi.org/10.21037/jmai-22-33","url":null,"abstract":"","PeriodicalId":73815,"journal":{"name":"Journal of medical artificial intelligence","volume":" ","pages":""},"PeriodicalIF":0.0,"publicationDate":"2023-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"46553171","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}

{"title":"Enabling scalable clinical interpretation of machine learning (ML)-based phenotypes using real world data","authors":"O. Parsons, N. Barlow, J. Baxter, K. Paraschin, Andrea Derix, Peter Hein, R. Dürichen","doi":"10.21037/jmai-22-42","DOIUrl":"https://doi.org/10.21037/jmai-22-42","url":null,"abstract":"","PeriodicalId":73815,"journal":{"name":"Journal of medical artificial intelligence","volume":"1 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2023-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"41562490","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}

{"title":"Artificial intelligence in screening for obstructive sleep apnoea syndrome (OSAS): a narrative review","authors":"Bei Pei, Ming Xia, Hong Jiang","doi":"10.21037/jmai-22-79","DOIUrl":"https://doi.org/10.21037/jmai-22-79","url":null,"abstract":"","PeriodicalId":73815,"journal":{"name":"Journal of medical artificial intelligence","volume":" ","pages":""},"PeriodicalIF":0.0,"publicationDate":"2023-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"42763342","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}

{"title":"Using a self-attention architecture to automate valence categorization of French teenagers' free descriptions of their family relationships. A proof of concept.","authors":"M. Sedki, N. Vidal, P. Roux, C. Barry, M. Speranza, B. Falissard, E. Brunet-Gouet","doi":"10.1101/2023.01.16.23284557","DOIUrl":"https://doi.org/10.1101/2023.01.16.23284557","url":null,"abstract":"This paper proposes a proof of concept of using natural language processing techniques to categorize valence of family relationships described in free texts written by french teenagers. The proposed study traces the evolution of techniques for word embedding. After decomposing the different texts in our possession into short texts composed of sentences and manual labeling, we tested different word embedding scenarios to train a multi-label classification model where a text can take several labels: labels describing the family link between the teenager and the person mentioned in the text and labels describing the teenager's relationship with them positive/negative/neutral valence). The natural baseline for word vector representation of our texts is to build a TF-IDF and train classical classifiers (Elasticnet logistic regression, gradient boosting, random forest, support vector classifier) after selecting a model by cross validation in each class of machine learning models. We then studied the strengths of word-vectors embeddings by an advanced language representation technique via the CamemBERT transformer model, and, again, used them with classical classifiers to compare their respective performances. The last scenario consisted in augmenting the CamemBERT with output dense layers (perceptron) representing a classifier adapted to the multi-label classification and fine-tuning the CamemBERT original layers. The optimal fine-tuning depth that achieves a bias-variance trade-off was obtained by a cross-validation procedure. The results of the comparison of the three scenarios on a test dataset show a clear improvement of the classification performances of the scenario with fine-tuning beyond the baseline and of a simple vectorization using CamemBERT without fine-tuning. Despite the moderate size of the dataset and the input texts, fine-tuning to an optimal depth remains the best solution to build a classifier.","PeriodicalId":73815,"journal":{"name":"Journal of medical artificial intelligence","volume":" ","pages":""},"PeriodicalIF":0.0,"publicationDate":"2023-01-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"45085579","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}

{"title":"Implementing artificial intelligence in clinical practice: a mixed-method study of barriers and facilitators","authors":"B. Schouten, M. Schinkel, A. W. Boerman, Petra van Pijkeren, Maureen Thodé, M. V. van Beneden, R. N. Nannan Panday, R. de Jonge, W. Wiersinga, P. Nanayakkara","doi":"10.21037/jmai-22-71","DOIUrl":"https://doi.org/10.21037/jmai-22-71","url":null,"abstract":"","PeriodicalId":73815,"journal":{"name":"Journal of medical artificial intelligence","volume":" ","pages":""},"PeriodicalIF":0.0,"publicationDate":"2022-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"47724594","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}

{"title":"Deep learning applications in coronary anatomy imaging: a systematic review and meta-analysis.","authors":"Ebraham Alskaf, Utkarsh Dutta, Cian M Scannell, Amedeo Chiribiri","doi":"10.21037/jmai-22-36","DOIUrl":"10.21037/jmai-22-36","url":null,"abstract":"<p><strong>Background: </strong>The application of deep learning on medical imaging is growing in prevalence in the recent literature. One of the most studied areas is coronary artery disease (CAD). Imaging of coronary artery anatomy is fundamental, which has led to a high number of publications describing a variety of techniques. The aim of this systematic review is to review the evidence behind the accuracy of deep learning applications in coronary anatomy imaging.</p><p><strong>Methods: </strong>The search for the relevant studies, which applied deep learning on coronary anatomy imaging, was performed in a systematic approach on MEDLINE and EMBASE databases, followed by reviewing of abstracts and full texts. The data from the final studies was retrieved using data extraction forms. A meta-analysis was performed on a subgroup of studies, which looked at fractional flow reserve (FFR) prediction. Heterogeneity was tested using tau², I² and Q tests. Finally, a risk of bias was performed using Quality Assessment of Diagnostic Accuracy Studies (QUADAS) approach.</p><p><strong>Results: </strong>A total of 81 studies met the inclusion criteria. The most common imaging modality was coronary computed tomography angiography (CCTA) (58%) and the most common deep learning method was convolutional neural network (CNN) (52%). The majority of studies demonstrated good performance metrics. The most common outputs were focused on coronary artery segmentation, clinical outcome prediction, coronary calcium quantification and FFR prediction, and most studies reported area under the curve (AUC) of ≥80%. The pooled diagnostic odds ratio (DOR) derived from 8 studies looking at FFR prediction using CCTA was 12.5 using the Mantel-Haenszel (MH) method. There was no significant heterogeneity amongst studies according to Q test (P=0.2496).</p><p><strong>Conclusions: </strong>Deep learning has been used in many applications on coronary anatomy imaging, most of which are yet to be externally validated and prepared for clinical use. The performance of deep learning, especially CNN models, proved to be powerful and some applications have already translated into medical practice, such as computed tomography (CT)-FFR. These applications have the potential to translate technology into better care of CAD patients.</p>","PeriodicalId":73815,"journal":{"name":"Journal of medical artificial intelligence","volume":"5 ","pages":"11"},"PeriodicalIF":0.0,"publicationDate":"2022-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ftp.ncbi.nlm.nih.gov/pub/pmc/oa_pdf/b7/84/EMS163415.PMC7614252.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"10826937","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}

{"title":"Sleep’s depth detection using electroencephalogram signal processing and neural network classification","authors":"M. Touil, L. Bahatti, A. El Magri","doi":"10.21037/jmai-22-32","DOIUrl":"https://doi.org/10.21037/jmai-22-32","url":null,"abstract":"","PeriodicalId":73815,"journal":{"name":"Journal of medical artificial intelligence","volume":"1 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2022-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"44602407","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}

{"title":"Accuracy of predicting IgHV mutation status in chronic lymphocytic leukemia using RNA expression profiling and machine learning","authors":"A. Charifa, Hong Zhang, A. Pecora, A. Ip, I. De Dios, Wanlong Ma, L. Leslie, T. Feldman, A. Goy, M. Albitar","doi":"10.21037/jmai-22-28","DOIUrl":"https://doi.org/10.21037/jmai-22-28","url":null,"abstract":"","PeriodicalId":73815,"journal":{"name":"Journal of medical artificial intelligence","volume":" ","pages":""},"PeriodicalIF":0.0,"publicationDate":"2022-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"47427615","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}

{"title":"A narrative review on radiotherapy practice in the era of artificial intelligence: how relevant is the medical physicist?","authors":"Eric Naab Manson, A. N. Mumuni, E. Fiagbedzi, I. Shirazu, H. Sulemana","doi":"10.21037/jmai-22-27","DOIUrl":"https://doi.org/10.21037/jmai-22-27","url":null,"abstract":"Background and Objective: Artificial intelligence (AI) uses computers and machines to simulate how the human mind makes decisions and solves problems. In radiotherapy practice, AI technologies continue to be promising in image registration, synthetic computed tomography (CT), image segmentation, motion management, treatment planning, and delivery procedures, patient follow-up and quality assurance (QA). This, therefore, provides a new window of opportunity to improve upon the accuracy and output times of the manual implementation of these procedures. The goal of this review was to explore how machine learning AI technologies in radiotherapy could affect the clinical practice of medical physicists. Methods: A narrative literature review was conducted from PubMed, Science Direct and Scopus using the search terms: in the English language within 6 months. Key Content and Findings: The roles of AI and the clinical medical physicist are complementary in radiotherapy practice. Both the medical physicists and AI technology are highly needed to support the full implementation and optimization of radiotherapy procedures. Conclusions: To achieve successful implementation of AI in radiotherapy and optimize radiotherapy procedures, clinical medical physicist should receive some compulsory training in AI technologies during their education and training. They should ultimately be involved in the incorporation of machine learning technologies in radiotherapy equipment. patient-specific dosimetric of patient treatment Dosimetric measurements in phantoms one of following detectors: portal imaging The third type of looks for delivery errors in log files generated during during delivery time-series linear quality","PeriodicalId":73815,"journal":{"name":"Journal of medical artificial intelligence","volume":" ","pages":""},"PeriodicalIF":0.0,"publicationDate":"2022-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"47113877","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}