Image-based Deep Learning Research Articles

Image-based deep learning in diagnosing mycoplasma pneumonia on pediatric chest X-rays.

Correctly diagnosing and accurately distinguishing mycoplasma pneumonia in children has consistently posed a challenge in clinical practice, as it can directly impact the prognosis of affected children. To address this issue, we analyzed chest X-rays (CXR) using various deep learning models to diagnose pediatric mycoplasma pneumonia. We collected 578 cases of children with mycoplasma infection and 191 cases of children with virus infection, with available CXR sets. Three deep convolutional neural networks (ResNet50, DenseNet121, and EfficientNetv2-S) were used to distinguish mycoplasma pneumonia from viral pneumonia based on CXR. Accuracy, area under the curve (AUC), sensitivity, and specificity were used to evaluate the performance of the model. Visualization was also achieved through the use of Class Activation Mapping (CAM), providing more transparent and interpretable classification results. Of the three models evaluated, ResNet50 outperformed the others. Pretrained with the ZhangLabData dataset, the ResNet50 model achieved 80.00% accuracy in the validation set. The model also showed robustness in two test sets, with accuracy of 82.65 and 83.27%, and AUC values of 0.822 and 0.758. In the test results using ImageNet pre-training weights, the accuracy of the ResNet50 model in the validation set was 80.00%; the accuracy in the two test sets was 81.63 and 62.91%; and the corresponding AUC values were 0.851 and 0.776. The sensitivity values were 0.884 and 0.595, and the specificity values were 0.655 and 0.814. This study demonstrates that deep convolutional networks utilizing transfer learning are effective in detecting mycoplasma pneumonia based on chest X-rays (CXR). This suggests that, in the near future, such computer-aided detection approaches can be employed for the early screening of pneumonia pathogens. This has significant clinical implications for the rapid diagnosis and appropriate medical intervention of pneumonia, potentially enhancing the prognosis for affected children.

A modified deep learning method for Alzheimer’s disease detection based on the facial submicroscopic features in mice

Alzheimer’s disease (AD) is a chronic disease among people aged 65 and older. As the aging population continues to grow at a rapid pace, AD has emerged as a pressing public health issue globally. Early detection of the disease is important, because increasing evidence has illustrated that early diagnosis holds the key to effective treatment of AD. In this work, we developed and refined a multi-layer cyclic Residual convolutional neural network model, specifically tailored to identify AD-related submicroscopic characteristics in the facial images of mice. Our experiments involved classifying the mice into two distinct groups: a normal control group and an AD group. Compared with the other deep learning models, the proposed model achieved a better detection performance in the dataset of the mouse experiment. The accuracy, sensitivity, specificity and precision for AD identification with our proposed model were as high as 99.78%, 100%, 99.65% and 99.44%, respectively. Moreover, the heat maps of AD correlation in the facial images of the mice were acquired with the class activation mapping algorithm. It was proven that the facial images contained AD-related submicroscopic features. Consequently, through our mouse experiments, we validated the feasibility and accuracy of utilizing a facial image-based deep learning model for AD identification. Therefore, the present study suggests the potential of using facial images for AD detection in humans through deep learning-based methods.

A comparative study on machine learning approaches for rock mass classification using drilling data

Current rock engineering design in drill and blast tunnelling primarily relies on engineers' observational assessments. Measure While Drilling (MWD) data, a high-resolution sensor dataset collected during tunnel excavation, is underutilised, mainly serving for geological visualisation. This study aims to automate the translation of MWD data into actionable metrics for rock engineering. It seeks to link data to specific engineering actions, thus providing critical decision support for geological challenges ahead of the tunnel face. Leveraging a large and geologically diverse dataset of ∼500,000 drillholes from 15 tunnels, the research introduces models for accurate rock mass quality classification in a real-world tunnelling context. Both conventional machine learning and image-based deep learning are explored to classify MWD data into Q-classes and Q-values—examples of metrics describing the stability of the rock mass—using both tabular- and image data. The results indicate that the K-nearest neighbours algorithm in an ensemble with tree-based models using tabular data effectively classifies rock mass quality. It achieves a cross-validated balanced accuracy of 0.86 in classifying rock mass into the Q-classes A, B, C, D, E1, E2, and 0.95 for a binary classification with E versus the rest. Classification using a CNN with MWD-images for each blasting round resulted in a balanced accuracy of 0.82 for binary classification. Regressing the Q-value from tabular MWD-data achieved cross-validated R2 and MSE scores of 0.80 and 0.18 for a similar ensemble model as in classification. High performance in regression and classification boosts confidence in automated rock mass assessment. Applying advanced modelling on a unique dataset demonstrates MWD data's value in improving rock mass classification accuracy and advancing data-driven rock engineering design, reducing manual intervention.

Image-based deep learning model to predict stoma-site incisional hernia in patients with temporary ileostomy: A retrospective study

Image-based deep learning model to predict stoma-site incisional hernia in patients with temporary ileostomy: A retrospective study

Deep learning assessment of senescence-associated nuclear morphologies in mammary tissue from healthy female donors to predict future risk of breast cancer: a retrospective cohort study

Cellular senescence has been associated with cancer as either a barrier mechanism restricting autonomous cell proliferation or a tumour-promoting microenvironmental mechanism that secretes proinflammatory paracrine factors. With most work done in non-human models and the heterogeneous nature of senescence, the precise role of senescent cells in the development of cancer in humans is not well understood. Furthermore, more than 1 million non-malignant breast biopsies are taken every year that could be a major resource for risk stratification. We aimed to explore the clinical relevance for breast cancer development of markers of senescence in mammary tissue from healthy female donors. In this retrospective cohort study, we applied single-cell deep learning senescence predictors, based on nuclear morphology, to histological images of haematoxylin and eosin-stained breast biopsy samples from healthy female donors at the Komen Tissue Bank (KTB) at the Indiana University Simon Cancer Center (Indianapolis, IN, USA). All KTB participants (aged ≥18 years) who underwent core biopsies for research purposes between 2009 and 2019 were eligible for the study. Senescence was predicted in the epithelial (terminal duct lobular units [TDLUs] and non-TDLU epithelium), stromal, and adipose tissue compartments using validated models, previously trained on cells induced to senescence by ionising radiation (IR), replicative exhaustion (or replicative senescence; RS), or antimycin A, atazanavir-ritonavir, and doxorubicin (AAD) exposures. To benchmark our senescence-based cancer prediction results, we generated 5-year Gail scores-the current clinical gold standard for breast cancer risk prediction-for participants aged 35 years and older on the basis of characteristics at the time of tissue donation. The primary outcome was estimated odds of breast cancer via logistic modelling for each tissue compartment based on predicted senescence scores in cases (participants who had been diagnosed with breast cancer as of data cutoff, July 31, 2022) and controls (those who had not been diagnosed with breast cancer). 4382 female donors (median age at donation 45 years [IQR 34-57]) were eligible for the study. As of data cutoff (median follow-up of 10 years [7-11]), 86 (2·0%) had developed breast cancer a mean of 4·8 years (SD 2·84) after date of donation and 4296 (98·0%) had not received a breast cancer diagnosis. Among the 86 cases, we found significant differences in adipose-specific IR and AAD senescence prediction scores compared with controls. Risk analysis showed that individuals in the upper half (above the median) of scores for the adipose tissue IR model had higher odds of developing breast cancer (odds ratio [OR] 1·71 [95% CI 1·10-2·68]; p=0·019), whereas the adipose AAD model revealed a reduced odds of developing breast cancer (OR 0·57 [0·36-0·88]; p=0·013). For the other tissue compartments and the RS model, no significant associations were found (except for stromal tissue via the IR model, had higher odds of developing breast cancer [OR 1·59, 1·03-2·49]). Individuals with both of the adipose risk factors had an OR of 3·32 (1·68-7·03; p=0·0009). Participants with 5-year Gail scores above the median had an OR for development of cancer of 2·33 (1·46-3·82; p=0·0012) compared with those with scores below the median. When combining Gail scores with our adipose AAD risk model, we found that individuals with both of these predictors had an OR of 4·70 (2·29-10·90; p<0·0001). When combining the Gail score with our adipose IR model, we found that individuals with both predictors had an OR of 3·45 (1·77-7·24; p=0·0002). Assessment of senescence-associated nuclear morphologies with deep learning allows prediction of future cancer risk from normal breast biopsy samples. The combination of multiple models improved prediction of future breast cancer compared with the current clinical benchmark, the Gail model. Our results suggest an important role for microscope image-based deep learning models in predicting future cancer development. Such models could be incorporated into current breast cancer risk assessment and screening protocols. Novo Nordisk Foundation, Danish Cancer Society, and the US National Institutes of Health.

Accuracy of deep learning in the differential diagnosis of coronary artery stenosis: a systematic review and meta-analysis

BackgroundIn recent years, as deep learning has received widespread attention in the field of heart disease, some studies have explored the potential of deep learning based on coronary angiography (CAG) or coronary CT angiography (CCTA) images in detecting the extent of coronary artery stenosis. However, there is still a lack of a systematic understanding of its diagnostic accuracy, impeding the advancement of intelligent diagnosis of coronary artery stenosis. Therefore, we conducted this study to review the accuracy of image-based deep learning in detecting coronary artery stenosis.MethodsWe retrieved PubMed, Cochrane, Embase, and Web of Science until April 11, 2023. The risk of bias in the included studies was appraised using the QUADAS-2 tool. We extracted the accuracy of deep learning in the test set and performed subgroup analyses by binary and multiclass classification scenarios. We performed a subgroup analysis based on different degrees of stenosis and applied a double arcsine transformation to process the data. The analysis was done by using R.ResultsOur systematic review finally included 18 studies, involving 3568 patients and 13,362 images. In the included studies, deep learning models were constructed based on CAG and CCTA. In binary classification tasks, the accuracy for detecting > 25%, > 50% and > 70% degrees of stenosis at the vessel level were 0.81 (95% CI: 0.71–0.85), 0.73 (95% CI: 0.58–0.88) and 0.61 (95% CI: 0.56–0.65), respectively. In multiclass classification tasks, the accuracy for detecting 0–25%, 25–50%, 50–70%, and 70–100% degrees of stenosis at the vessel level were 0.78 (95% CI: 0.73–0.84), 0.86 (95% CI: 0.78–0.93), 0.83 (95% CI: 0.70–0.97), and 0.70 (95% CI: 0.42–0.98), respectively.ConclusionsOur study shows that deep learning models based on CAG and CCTA appear to be relatively accurate in diagnosing different degrees of coronary artery stenosis. However, for various degrees of stenosis, their accuracy still needs to be further improved.

189P An image-based deep learning prediction model for characterization of the drug tolerant persister cell state

189P An image-based deep learning prediction model for characterization of the drug tolerant persister cell state

IMAGE-BASED YOLOV4 ARCHITECTURE FOR DETECTING MINERAL IN SEDIMENTARY ROCKS THIN SECTION

Thin section analysis of sedimentary rocks is the basis for identifying minerals and textures. In general, quantitative analysis of thin sections of rock often requires many hours of work when done manually. In today's era, mineralogical interpretation and percentage calculations must be carried out automatically using more practical applications. The research method begins with the identification of 44 thin section samples in parallel plane polarized (PPL) and crossed polarized (XPL) conditions with thin section analysis then mineralogy detection is carried out using a computational approach, namely the use of image-based Deep Learning YoloV4 architecture with 2D RGB image objects from the thin section of sedimentary rock. The results of this study show the best values of Average Precision in Quartz, Feldspar, and lithic are 39.21% in the XPL model, 26.53% in the XPL model, and 15.75% in combined mode, according to the training and testing of YoloV4 Models for the identification of rock minerals in thin sections. Based on the complexity of the mineral types, the granularity of the detection, and the specific geological objectives, establishing a meaningful benchmark or baseline for comparison is always challenging. Additionally, consider discussing the trade-offs between precision and recall, as a higher precision may be more critical in some geological applications. It is expected that the application of this research can produce practical, fast and accurate interpretation of the determination of minerals in sedimentary rocks from all thin-section images of rocks and thus provide a complete understanding of geological views automatically.

AI-powered visual diagnosis of vulvar lichen sclerosus: A pilot study.

Vulvar lichen sclerosus (VLS) is a chronic inflammatory skin condition associated with significant impairment of quality of life and potential risk of malignant transformation. However, diagnosis of VLS is often delayed due to its variable clinical presentation and shame-related late consultation. Machine learning (ML)-trained image recognition software could potentially facilitate early diagnosis of VLS. To develop a ML-trained image-based model for the detection of VLS. Images of both VLS and non-VLS anogenital skin were collected, anonymized, and selected. In the VLS images, 10 typical skin signs (whitening, hyperkeratosis, purpura/ecchymosis, erosion/ulcers/excoriation, erythema, labial fusion, narrowing of the introitus, labia minora resorption, posterior commissure (fourchette) band formation and atrophic shiny skin) were manually labelled. A deep convolutional neural network was built using the training set as input data and then evaluated using the test set, where the developed algorithm was run three times and the results were then averaged. A total of 684 VLS images and 403 non-VLS images (70% healthy vulva and 30% with other vulvar diseases) were included after the selection process. A deep learning algorithm was developed by training on 775 images (469 VLS and 306 non-VLS) and testing on 312 images (215 VLS and 97 non-VLS). This algorithm performed accurately in discriminating between VLS and non-VLS cases (including healthy individuals and non-VLS dermatoses), with mean values of 0.94, 0.99 and 0.95 for recall, precision and accuracy, respectively. This pilot project demonstrated that our image-based deep learning model can effectivelydiscriminate between VLS and non-VLS skin, representing a promising tool for future use by clinicians and possibly patients. However, prospective studies are needed to validate the applicability and accuracy of our model in a real-world setting.

Clinical and dermoscopy image-based deep learning models for skin lesion diagnosis in clinical practice

Clinical and dermoscopy image-based deep learning models for skin lesion diagnosis in clinical practice

Exploring the Frontiers of Synthetic Image-Based Deep Learning Training in Digital X-ray Radiography

Exploring the Frontiers of Synthetic Image-Based Deep Learning Training in Digital X-ray Radiography

Integrated image-based deep learning and language models for primary diabetes care

Primary diabetes care and diabetic retinopathy (DR) screening persist as major public health challenges due to a shortage of trained primary care physicians (PCPs), particularly in low-resource settings. Here, to bridge the gaps, we developed an integrated image–language system (DeepDR-LLM), combining a large language model (LLM module) and image-based deep learning (DeepDR-Transformer), to provide individualized diabetes management recommendations to PCPs. In a retrospective evaluation, the LLM module demonstrated comparable performance to PCPs and endocrinology residents when tested in English and outperformed PCPs and had comparable performance to endocrinology residents in Chinese. For identifying referable DR, the average PCP’s accuracy was 81.0% unassisted and 92.3% assisted by DeepDR-Transformer. Furthermore, we performed a single-center real-world prospective study, deploying DeepDR-LLM. We compared diabetes management adherence of patients under the unassisted PCP arm (n = 397) with those under the PCP+DeepDR-LLM arm (n = 372). Patients with newly diagnosed diabetes in the PCP+DeepDR-LLM arm showed better self-management behaviors throughout follow-up (P < 0.05). For patients with referral DR, those in the PCP+DeepDR-LLM arm were more likely to adhere to DR referrals (P < 0.01). Additionally, DeepDR-LLM deployment improved the quality and empathy level of management recommendations. Given its multifaceted performance, DeepDR-LLM holds promise as a digital solution for enhancing primary diabetes care and DR screening.

Deep learning model based on endoscopic images predicting treatment response in locally advanced rectal cancer undergo neoadjuvant chemoradiotherapy: a multicenter study

PurposeNeoadjuvant chemoradiotherapy has been the standard practice for patients with locally advanced rectal cancer. However, the treatment response varies greatly among individuals, how to select the optimal candidates for neoadjuvant chemoradiotherapy is crucial. This study aimed to develop an endoscopic image-based deep learning model for predicting the response to neoadjuvant chemoradiotherapy in locally advanced rectal cancer.MethodsIn this multicenter observational study, pre-treatment endoscopic images of patients from two Chinese medical centers were retrospectively obtained and a deep learning-based tumor regression model was constructed. Treatment response was evaluated based on the tumor regression grade and was defined as good response and non-good response. The prediction performance of the deep learning model was evaluated in the internal and external test sets. The main outcome was the accuracy of the treatment prediction model, measured by the AUC and accuracy.ResultsThis deep learning model achieved favorable prediction performance. In the internal test set, the AUC and accuracy were 0.867 (95% CI: 0.847–0.941) and 0.836 (95% CI: 0.818–0.896), respectively. The prediction performance was fully validated in the external test set, and the model had an AUC of 0.758 (95% CI: 0.724–0.834) and an accuracy of 0.807 (95% CI: 0.774–0.843).ConclusionThe deep learning model based on endoscopic images demonstrated exceptional predictive power for neoadjuvant treatment response, highlighting its potential for guiding personalized therapy.

O-025 An artificial intelligence (AI) model non-invasively evaluates endometrial receptivity from ultrasound images, surpassing endometrial thickness (EMT) in predicting implantation

Abstract Study question Can an AI image-based model be trained to predict whether an endometrium, prior to progesterone, is receptive for successful embryo implantation from ultrasound images? Summary answer An endometrial receptivity AI model, utilizing ultrasound images and clinical features, is predictive of successful implantation and outperforms endometrial thickness as a predictor of implantation. What is known already Despite its importance in successful embryo implantation and subsequent live birth, endometrial receptivity is difficult to measure. The standard of care at some clinics is to cancel an embryo transfer if EMT is measuring less than 7 mm on ultrasound assessment – however, achieving an EMT threshold does not guarantee success, even with euploid blastocysts. Currently available receptivity tests (e.g. endometrial receptivity assays) are unreliable and have not shown significant improvements in reproductive outcomes, necessitating the development of new evaluation methods for endometrial receptivity. AI can potentially address this gap in clinical care and understanding, aiming to increase patient success. Study design, size, duration 79,602 ultrasound images (40,910 patients) of the endometrium, on the day of progesterone start according to patient’s frozen embryo transfer cycle, were retrospectively collected from 4 clinic networks including 70 clinic locations. PGT-A outcomes and blastocyst quality were available for 15,466 and 36,070 blastocysts respectively, with 14,538 having labels for both. Different image samples (based on embryo quality) were selected to train and test several AI models, with implantation (positive beta-hCG) as a binary outcome. Participants/materials, setting, methods Three criteria for negative (implantation) class sample selection were tested for receptivity model development: 1) high- or medium-quality and/or euploid blastocysts; 2) high-quality blastocysts; 3) all blastocysts. Models were ensembles consisting of an image-based deep learning (DL)-model and a feature-based machine learning (ML)-model. Model performance was measured by Area-under-the-curve (AUC), sensitivity, and specificity. Feature importance from the ML model was assessed, while receptivity model performance was compared to the power of EMT to predict implantation. Main results and the role of chance A receptivity model trained on scenario 1) (n = 27,424) achieved AUC 0.613, sensitivity 0.716, and specificity 0.439 on a test set of 9197 samples, in predicting implantation. Another receptivity model trained on scenario 2) (n = 25,625) achieved AUC 0.627, sensitivity 0.637, and specificity 0.543 on a test set of 8642. Whereas a third receptivity model trained on scenario 3) (n = 31,238) was selected for further analysis as it attained the best performance with AUC 0.631, sensitivity 0.628, and specificity 0.556 on a test set of 10,422. The ensemble model also utilizes relevant clinical features, which were ranked by importance to model predictions: EMT; progesterone blood test value; age at transfer; previous total embryo transfers; days between ultrasound date and transfer date; age at retrieval; and oocyte origin. EMT was assessed for its ability to predict implantation at thresholds between 5–20 mm. The threshold that best separates positive and negative implantation was determined by evaluating AUC on a validation set of 5682. A threshold of 8.8 mm achieved an AUC 0.576, sensitivity 0.694, and specificity 0.455 in predicting implantation on a test set of 5686. However, the receptivity model AUC (0.631) was significantly higher (p &amp;lt; 0.001; DeLong test for differences in AUC), outperforming EMT predictability. Limitations, reasons for caution The endometrial receptivity model was built and tested on retrospective frozen embryo transfer data. Results should be corroborated in prospective and non-selection studies, as well as in multiple geographical locations (all clinics included reside in the United States). The dataset requires greater diversification, with a greater representation of EMT (&amp;lt;7mm). Wider implications of the findings This study indicates that an AI model can be built to predict implantation from ultrasound images and clinical features, surpassing current standards - EMT alone. Further experimentation, with additional clinical features, may improve performance to facilitate development of a model that can accurately predict the gold-standard of live birth. Trial registration number Not applicable

Selection of Convolutional Neural Network Model for Bladder Tumor Classification of Cystoscopy Images and Comparison with Humans.

Purpose: An investigation of various convolutional neural network (CNN)-based deep learning algorithms was conducted to select the appropriate artificial intelligence (AI) model for calculating the diagnostic performance of bladder tumor classification on cystoscopy images, with the performance of the selected model to be compared against that of medical students and urologists. Methods: A total of 3,731 cystoscopic images that contained 2,191 tumor images were obtained from 543 bladder tumor cases and 219 normal cases were evaluated. A total of 17 CNN models were trained for tumor classification with various hyperparameters. The diagnostic performance of the selected AI model was compared with the results obtained from urologists and medical students by using the receiver operating characteristic (ROC) curve graph and metrics. Results: EfficientNetB0 was selected as the appropriate AI model. In the test results, EfficientNetB0 achieved a balanced accuracy of 81%, sensitivity of 88%, specificity of 74%, and an area under the curve (AUC) of 92%. In contrast, human-derived diagnostic statistics for the test data showed an average balanced accuracy of 75%, sensitivity of 94%, and specificity of 55%. Specifically, urologists had an average balanced accuracy of 91%, sensitivity of 95%, and specificity of 88%, while medical students had an average balanced accuracy of 69%, sensitivity of 94%, and specificity of 44%. Conclusions: Among the various AI models, we suggest that EfficientNetB0 is an appropriate AI classification model for determining the presence of bladder tumors in cystoscopic images. EfficientNetB0 showed the highest performance among several models and showed high accuracy and specificity compared to medical students. This AI technology will be helpful for less experienced urologists or nonurologists in making diagnoses. Image-based deep learning classifies bladder cancer using cystoscopy images and shows promise for generalized applications in biomedical image analysis and clinical decision making.

Mixed Supervision of Histopathology Improves Prostate Cancer Classification From MRI.

Non-invasive prostate cancer classification from MRI has the potential to revolutionize patient care by providing early detection of clinically significant disease, but has thus far shown limited positive predictive value. To address this, we present a image-based deep learning method to predict clinically significant prostate cancer from screening MRI in patients that subsequently underwent biopsy with results ranging from benign pathology to the highest grade tumors. Specifically, we demonstrate that mixed supervision via diverse histopathological ground truth improves classification performance despite the cost of reduced concordance with image-based segmentation. Where prior approaches have utilized pathology results as ground truth derived from targeted biopsies and whole-mount prostatectomy to strongly supervise the localization of clinically significant cancer, our approach also utilizes weak supervision signals extracted from nontargeted systematic biopsies with regional localization to improve overall performance. Our key innovation is performing regression by distribution rather than simply by value, enabling use of additional pathology findings traditionally ignored by deep learning strategies. We evaluated our model on a dataset of 973 (testing n=198 ) multi-parametric prostate MRI exams collected at UCSF from 2016-2019 followed by MRI/ultrasound fusion (targeted) biopsy and systematic (nontargeted) biopsy of the prostate gland, demonstrating that deep networks trained with mixed supervision of histopathology can feasibly exceed the performance of the Prostate Imaging-Reporting and Data System (PI-RADS) clinical standard for prostate MRI interpretation (71.6% vs 66.7% balanced accuracy and 0.724 vs 0.716 AUC).

Association of retinal image-based, deep learning cardiac BioAge with telomere length and cardiovascular biomarkers.

Our retinal image-based deep learning (DL) cardiac biological age (BioAge) model could facilitate fast, accurate, noninvasive screening for cardiovascular disease (CVD) in novel community settings and thus improve outcome with those with limited access to health care services. This study aimed to determine whether the results issued by our DL cardiac BioAge model are consistent with the known trends of CVD risk and the biomarker leukocyte telomere length (LTL), in a cohort of individuals from the UK Biobank. A cross-sectional cohort study was conducted using those individuals in the UK Biobank who had LTL data. These individuals were divided by sex, ranked by LTL, and then grouped into deciles. The retinal images were then presented to the DL model, and individual's cardiac BioAge was determined. Individuals within each LTL decile were then ranked by cardiac BioAge, and the mean of the CVD risk biomarkers in the top and bottom quartiles was compared. The relationship between an individual's cardiac BioAge, the CVD biomarkers, and LTL was determined using traditional correlation statistics. The DL cardiac BioAge model was able to accurately stratify individuals by the traditional CVD risk biomarkers, and for both males and females, those issued with a cardiac BioAge in the top quartile of their chronological peer group had a significantly higher mean systolic blood pressure, hemoglobin A 1c , and 10-year Pooled Cohort Equation CVD risk scores compared with those individuals in the bottom quartile (p<0.001). Cardiac BioAge was associated with LTL shortening for both males and females (males: -0.22, r2 = 0.04; females: -0.18, r2 = 0.03). In this cross-sectional cohort study, increasing CVD risk whether assessed by traditional biomarkers, CVD risk scoring, or our DL cardiac BioAge, CVD risk model, was inversely related to LTL. At a population level, our data support the growing body of evidence that suggests LTL shortening is a surrogate marker for increasing CVD risk and that this risk can be captured by our novel DL cardiac BioAge model.

Image-based deep learning model using DNA methylation data predicts the origin of cancer of unknown primary

Cancer of unknown primary (CUP) is a rare type of metastatic cancer in which the origin of the tumor is unknown. Since the treatment strategy for patients with metastatic tumors depends on knowing the primary site, accurate identification of the origin site is important. Here, we developed an image-based deep-learning model that utilizes a vision transformer algorithm for predicting the origin of CUP. Using DNA methylation dataset of 8,233 primary tumors from The Cancer Genome Atlas (TCGA), we categorized 29 cancer types into 18 organ classes and extracted 2,312 differentially methylated CpG sites (DMCs) from non-squamous cancer group and 420 DMCs from squamous cell cancer group. Using these DMCs, we created organ-specific DNA methylation images and used them for model training and testing. Model performance was evaluated using 394 metastatic cancer samples from TCGA (TCGA-meta) and 995 samples (693 primary and 302 metastatic cancers) obtained from 20 independent external studies. We identified that the DNA methylation image reveals a distinct pattern based on the origin of cancer. Our model achieved an overall accuracy of 96.95 % in the TCGA-meta dataset. In the external validation datasets, our classifier achieved overall accuracies of 96.39 % and 94.37 % in primary and metastatic tumors, respectively. Especially, the overall accuracies for both primary and metastatic samples of non-squamous cell cancer were exceptionally high, with 96.79 % and 96.85 %, respectively.

Accurately identifying the defects of bubbles and foreign objects under the protective films of electric vehicle batteries by using 3D point clouds

Abstract For the defects of bubbles and foreign objects under the protective film of electric vehicle batteries, it is difficult to accurately identify them over traditional 2D optical images. In this paper, we first propose a supervoxel-based region growing algorithm for pre-segmentation of point clouds. Secondly, we utilize radial basis function interpolation and threshold segmentation methods to accurately segment defect point clouds from the entire point cloud. Finally, we develop a feature descriptor and combine it with support vector machine to classify bubbles and foreign objects under the film. This paper achieves the identification of bubbles and foreign objects under the film through two steps: point cloud segmentation and point cloud classification. Experimental results demonstrate that the proposed point cloud segmentation method exhibits high robustness to noise and the intrinsic curvature of the workpiece. Additionally, in the classification scenario presented in this paper, the proposed feature descriptor outperforms classical feature descriptors. Compared to image-based deep learning methods, the defect recognition algorithm proposed in this paper has clear principles and superior performance, with precision and recall of 95.63% and 96.95%, and an intersection over union metric of 0.926.

PET/CT deep learning prognosis for treatment decision support in esophageal squamous cell carcinoma

ObjectivesThe clinical decision-making regarding choosing surgery alone (SA) or surgery followed by postoperative adjuvant chemotherapy (SPOCT) in esophageal squamous cell carcinoma (ESCC) remains controversial. We aim to propose a pre-therapy PET/CT image-based deep learning approach to improve the survival benefit and clinical management of ESCC patients.MethodsThis retrospective multicenter study included 837 ESCC patients from three institutions. Prognostic biomarkers integrating six networks were developed to build an ESCC prognosis (ESCCPro) model and predict the survival probability of ESCC patients treated with SA and SPOCT. Patients who did not undergo surgical resection were in a control group. Overall survival (OS) was the primary end-point event. The expected improvement in survival prognosis with the application of ESCCPro to assign treatment protocols was estimated by comparing the survival of patients in each subgroup. Seven clinicians with varying experience evaluated how ESCCPro performed in assisting clinical decision-making.ResultsIn this retrospective multicenter study, patients receiving SA had a median OS 9.2 months longer than controls. No significant differences in survival were found between SA patients with predicted poor outcomes and the controls (p > 0.05). It was estimated that if ESCCPro was used to determine SA and SPOCT eligibility, the median OS in the ESCCPro-recommended SA group and SPOCT group would have been 15.3 months and 24.9 months longer, respectively. In addition, ESCCPro also significantly improved prognosis accuracy, certainty, and the efficiency of clinical experts.ConclusionESCCPro assistance improved the survival benefit of ESCC patients and the clinical decision-making among the two treatment approaches.Critical relevance statementThe ESCCPro model for treatment decision-making is promising to improve overall survival in ESCC patients undergoing surgical resection and patients undergoing surgery followed by postoperative adjuvant chemotherapy.Key PointsESCC is associated with a poor prognosis and unclear ideal treatments.ESCCPro predicts the survival of patients with ESCC and the expected benefit from SA.ESCCPro improves clinicians’ stratification of patients’ prognoses.Graphical