scholarly journals Joint segmentation and classification of hepatic lesions in ultrasound images using deep learning

2021 ◽  
Author(s):  
Hwaseong Ryu ◽  
Seung Yeon Shin ◽  
Jae Young Lee ◽  
Kyoung Mu Lee ◽  
Hyo-jin Kang ◽  
...  
Keyword(s):  
Deep Learning ◽  
Classification System ◽  
Characteristic Curve ◽  
Input Image ◽  
Hepatic Lesion ◽  
Ultrasound Images ◽  
Joint System ◽  
The Mean ◽  
Hepatic Lesions ◽  
Sensitivity Specificity

Abstract Objectives To develop a convolutional neural network system to jointly segment and classify a hepatic lesion selected by user clicks in ultrasound images. Methods In total, 4309 anonymized ultrasound images of 3873 patients with hepatic cyst (n = 1214), hemangioma (n = 1220), metastasis (n = 1001), or hepatocellular carcinoma (HCC) (n = 874) were collected and annotated. The images were divided into 3909 training and 400 test images. Our network is composed of one shared encoder and two inference branches used for segmentation and classification and takes the concatenation of an input image and two Euclidean distance maps of foreground and background clicks provided by a user as input. The performance of hepatic lesion segmentation was evaluated based on the Jaccard index (JI), and the performance of classification was based on accuracy, sensitivity, specificity, and the area under the receiver operating characteristic curve (AUROC). Results We achieved performance improvements by jointly conducting segmentation and classification. In the segmentation only system, the mean JI was 68.5%. In the classification only system, the accuracy of classifying four types of hepatic lesions was 79.8%. The mean JI and classification accuracy were 68.5% and 82.2%, respectively, for the proposed joint system. The optimal sensitivity and specificity and the AUROC of classifying benign and malignant hepatic lesions of the joint system were 95.0%, 86.0%, and 0.970, respectively. The respective sensitivity, specificity, and the AUROC for classifying four hepatic lesions of the joint system were 86.7%, 89.7%, and 0.947. Conclusions The proposed joint system exhibited fair performance compared to segmentation only and classification only systems. Key Points • The joint segmentation and classification system using deep learning accurately segmented and classified hepatic lesions selected by user clicks in US examination. • The joint segmentation and classification system for hepatic lesions in US images exhibited higher performance than segmentation only and classification only systems. • The joint segmentation and classification system could assist radiologists with minimal experience in US imaging by characterizing hepatic lesions.

scholarly journals Automated Breast Cancer Detection in Digital Mammograms of Various Densities via Deep Learning

2020 ◽  
Vol 10 (4) ◽  
pp. 211 ◽  
Author(s):  
Yong Joon Suh ◽  
Jaewon Jung ◽  
Bum-Joo Cho
Keyword(s):  
Breast Cancer ◽  
Deep Learning ◽  
Operating Characteristic ◽  
Meta Analysis ◽  
Characteristic Curve ◽  
Malignant Lesion ◽  
Model Performance ◽  
Mean Values ◽  
The Mean ◽  
Deep Learning Model

Mammography plays an important role in screening breast cancer among females, and artificial intelligence has enabled the automated detection of diseases on medical images. This study aimed to develop a deep learning model detecting breast cancer in digital mammograms of various densities and to evaluate the model performance compared to previous studies. From 1501 subjects who underwent digital mammography between February 2007 and May 2015, craniocaudal and mediolateral view mammograms were included and concatenated for each breast, ultimately producing 3002 merged images. Two convolutional neural networks were trained to detect any malignant lesion on the merged images. The performances were tested using 301 merged images from 284 subjects and compared to a meta-analysis including 12 previous deep learning studies. The mean area under the receiver-operating characteristic curve (AUC) for detecting breast cancer in each merged mammogram was 0.952 ± 0.005 by DenseNet-169 and 0.954 ± 0.020 by EfficientNet-B5, respectively. The performance for malignancy detection decreased as breast density increased (density A, mean AUC = 0.984 vs. density D, mean AUC = 0.902 by DenseNet-169). When patients’ age was used as a covariate for malignancy detection, the performance showed little change (mean AUC, 0.953 ± 0.005). The mean sensitivity and specificity of the DenseNet-169 (87 and 88%, respectively) surpassed the mean values (81 and 82%, respectively) obtained in a meta-analysis. Deep learning would work efficiently in screening breast cancer in digital mammograms of various densities, which could be maximized in breasts with lower parenchyma density.

scholarly journals Can Deep Learning-Based Volumetric Analysis Predict Oxygen Demand Increase in Patients with COVID-19 Pneumonia?

Medicina ◽  
2021 ◽  
Vol 57 (11) ◽  
pp. 1148
Author(s):  
Marie Takahashi ◽  
Tomoyuki Fujioka ◽  
Toshihiro Horii ◽  
Koichiro Kimura ◽  
Mizuki Kimura ◽  
...  
Keyword(s):  
Deep Learning ◽  
Characteristic Curve ◽  
Oxygen Demand ◽  
Volumetric Analysis ◽  
Ct Scans ◽  
Image Analysis System ◽  
Volume Data ◽  
Respiratory Status ◽  
Analysis System ◽  
Sensitivity Specificity

Background and Objectives: This study aimed to investigate whether predictive indicators for the deterioration of respiratory status can be derived from the deep learning data analysis of initial chest computed tomography (CT) scans of patients with coronavirus disease 2019 (COVID-19). Materials and Methods: Out of 117 CT scans of 75 patients with COVID-19 admitted to our hospital between April and June 2020, we retrospectively analyzed 79 CT scans that had a definite time of onset and were performed prior to any medication intervention. Patients were grouped according to the presence or absence of increased oxygen demand after CT scan. Quantitative volume data of lung opacity were measured automatically using a deep learning-based image analysis system. The sensitivity, specificity, and area under the receiver operating characteristic curve (AUC) of the opacity volume data were calculated to evaluate the accuracy of the system in predicting the deterioration of respiratory status. Results: All 79 CT scans were included (median age, 62 years (interquartile range, 46–77 years); 56 (70.9%) were male. The volume of opacity was significantly higher for the increased oxygen demand group than for the nonincreased oxygen demand group (585.3 vs. 132.8 mL, p < 0.001). The sensitivity, specificity, and AUC were 76.5%, 68.2%, and 0.737, respectively, in the prediction of increased oxygen demand. Conclusion: Deep learning-based quantitative analysis of the affected lung volume in the initial CT scans of patients with COVID-19 can predict the deterioration of respiratory status to improve treatment and resource management.

scholarly journals Change Detection in Water reservoir using Image Fusion and Supervised Classifier

2021 ◽  
pp. 324-330
Author(s):  
Kiran Khandarkar ◽  
Dr. Sharvari Tamne
Keyword(s):  
Change Detection ◽  
Image Fusion ◽  
Classification System ◽  
Supervised Classification ◽  
Water Reservoir ◽  
Input Image ◽  
Sar Images ◽  
The Mean ◽  
Log Ratio

The research provides a method for improving change detection in SAR images using a fusion object and a supervised classification system. To remove noise from the input image, we use the DnCNN denoising approach. The data from the first image is then processed with the mean ratio operator. The log ratio operator is used to process the second image. These two images are fused together using SWT-based image fusion, and the output is sent to a supervise classifier for change detection.

Detection of Thyroid Nodules with Ultrasound Images Based on Deep Learning

2020 ◽  
Vol 16 (2) ◽  
pp. 174-180 ◽  
Author(s):  
Xia Yu ◽  
Hongjie Wang ◽  
Liyong Ma
Keyword(s):  
Deep Learning ◽  
Thyroid Nodule ◽  
Thyroid Nodules ◽  
Detection Method ◽  
Ultrasound Images ◽  
True Negative ◽  
Automated Method ◽  
High Incidence ◽  
Sensitivity Specificity ◽  
Physician Workload

Background: Thyroid nodules are a common clinical entity with high incidence. Ultrasound is often employed to detect and evaluate thyroid nodules. The development of an efficient automated method to detect thyroid nodules using ultrasound has the potential to reduce both physician workload and operator-dependence. Objective: To study the method of automatic detection of thyroid nodules based on deep learning using ultrasound, and to obtain the detection method with higher accuracy and better performance. Methods: A total of 1200 ultrasound images of thyroid nodules and 800 ultrasound thyroid images without nodule are collected. An improved faster R-CNN based detection method of thyroid nodule is proposed. Instead of using VGG16 as the backbone, ResNet is employed as the backbone for faster R-CNN. SVM, CNN and Faster-RCNN methods are used for thyroid nodule detection test. Precision, sensitivity, specificity and F1-score indicators are used to evaluate the detection performance of different methods. Results: The method based on deep learning is superior to that based on SVM. Faster R-CNN method and the improved method are better than CNN method. Compared with VGG16 as the backbone, RestNet101 backbone based faster R-CNN method achieves better thyroid detection effect. From the accuracy index, the proposed method is 0.084, 0.032 and 0.019 higher than SVM, CNN and faster R-CNN, respectively. Similar results can be seen in precision, sensitivity, specificity and F1-Score indicators. Conclusion: The proposed method of deep learning achieves the best performance values with the highest true positive and true negative detection compared to other methods and performs best in the detection of thyroid nodules.

scholarly journals Validation of the STOP-Bang questionnaire as a means of screening for obstructive sleep apnea in adults in Brazil

2017 ◽  
Vol 43 (6) ◽  
pp. 456-463 ◽  
Author(s):  
Ricardo Luiz de Menezes Duarte ◽  
Lorena Barbosa de Moraes Fonseca ◽  
Flavio José Magalhães-da-Silveira ◽  
Erika Aparecida da Silveira ◽  
Marcelo Fouad Rabahi
Keyword(s):  
Obstructive Sleep Apnea ◽  
Sleep Apnea ◽  
Characteristic Curve ◽  
Apnea Hypopnea Index ◽  
Language Version ◽  
Culturally Adapted ◽  
Obstructive Sleep ◽  
The Mean ◽  
And Gender ◽  
Sensitivity Specificity

ABSTRACT Objective: To validate the Portuguese-language version of the STOP-Bang (acronym for Snoring, Tiredness, Observed apnea, high blood Pressure, Body mass index, Age, Neck circumference, and Gender) questionnaire, culturally adapted for use in Brazil, as a means of screening for obstructive sleep apnea (OSA) in adults. Methods: In this validation study, we enrolled patients ≥ 18 years of age, recruited between May of 2015 and November of 2016. All patients completed the STOP-Bang questionnaire and underwent overnight polysomnography. To evaluate the performance of the questionnaire, we used contingency tables and areas under the (receiver operating characteristic) curve (AUCs). Results: We included 456 patients. The mean age was 43.7 ± 12.5 years, and 291 (63.8%) of the patients were male. On the basis of the apnea-hypopnea index (AHI), we categorized OSA as mild/moderate/severe (any OSA; AHI ≥ 5 events/h), moderate/severe (AHI ≥ 15 events/h), or severe (AHI ≥ 30 events/h). The overall prevalence of OSA was 78.3%, compared with 52.0%, and 28.5% for moderate/severe and severe OSA, respectively. The most common score on the STOP-Bang questionnaire was 4 points (n = 106), followed by 3 points (n = 85) and 5 points (n = 82). An increase in the score was paralleled by a reduction in sensitivity with a corresponding increase in specificity for all AHI cut-off points. The AUCs obtained for the identification of any, moderate/severe, and severe OSA were: 0.743, 0.731, and 0.779, respectively. For any OSA, the score on the questionnaire (cut-off, ≥ 3 points) presented sensitivity, specificity, and accuracy of 83.5%, 45.5%, and 75.2%, respectively. Conclusions: The STOP-Bang questionnaire performed adequately for OSA screening, indicating that it could be used as an effective screening tool for the disorder.

scholarly journals Development of a convolutional neural network to differentiate among the etiology of similar appearing pathological B lines on lung ultrasound: a deep learning study

BMJ Open ◽  
2021 ◽  
Vol 11 (3) ◽  
pp. e045120
Author(s):  
Robert Arntfield ◽  
Blake VanBerlo ◽  
Thamer Alaifan ◽  
Nathan Phelps ◽  
Matthew White ◽  
...  
Keyword(s):  
Neural Network ◽  
Deep Learning ◽  
Convolutional Neural Network ◽  
Lung Ultrasound ◽  
Low Cost ◽  
Characteristic Curve ◽  
Ultrasound Images ◽  
Diagnostic Specificity ◽  
Multicentre Research ◽  
B Lines

ObjectivesLung ultrasound (LUS) is a portable, low-cost respiratory imaging tool but is challenged by user dependence and lack of diagnostic specificity. It is unknown whether the advantages of LUS implementation could be paired with deep learning (DL) techniques to match or exceed human-level, diagnostic specificity among similar appearing, pathological LUS images.DesignA convolutional neural network (CNN) was trained on LUS images with B lines of different aetiologies. CNN diagnostic performance, as validated using a 10% data holdback set, was compared with surveyed LUS-competent physicians.SettingTwo tertiary Canadian hospitals.Participants612 LUS videos (121 381 frames) of B lines from 243 distinct patients with either (1) COVID-19 (COVID), non-COVID acute respiratory distress syndrome (NCOVID) or (3) hydrostatic pulmonary edema (HPE).ResultsThe trained CNN performance on the independent dataset showed an ability to discriminate between COVID (area under the receiver operating characteristic curve (AUC) 1.0), NCOVID (AUC 0.934) and HPE (AUC 1.0) pathologies. This was significantly better than physician ability (AUCs of 0.697, 0.704, 0.967 for the COVID, NCOVID and HPE classes, respectively), p<0.01.ConclusionsA DL model can distinguish similar appearing LUS pathology, including COVID-19, that cannot be distinguished by humans. The performance gap between humans and the model suggests that subvisible biomarkers within ultrasound images could exist and multicentre research is merited.

scholarly journals The value of S-Detect in improving the diagnostic performance of radiologists for the differential diagnosis of thyroid nodules

2020 ◽  
Vol 22 (4) ◽  
pp. 415
Author(s):  
Qi Wei ◽  
Shu-E Zeng ◽  
Li-Ping Wang ◽  
Yu-Jing Yan ◽  
Ting Wang ◽  
...  
Keyword(s):  
Diagnostic Accuracy ◽  
Predictive Value ◽  
Diagnostic Performance ◽  
Thyroid Nodules ◽  
Characteristic Curve ◽  
Diagnostic Value ◽  
Ultrasound Images ◽  
Experienced Radiologist ◽  
Before And After ◽  
Sensitivity Specificity

Aims: To compare the diagnostic value of S-Detect (a computer aided diagnosis system using deep learning) in differentiating thyroid nodules in radiologists with different experience and to assess if S-Detect can improve the diagnostic performance of radiologists.Materials and methods: Between February 2018 and October 2019, 204 thyroid nodules in 181 patients were included. An experienced radiologist performed ultrasound for thyroid nodules and obtained the result of S-Detect. Four radiologists with different experience on thyroid ultrasound (Radiologist 1, 2, 3, 4 with 1, 4, 9, 20 years, respectively) analyzed the conventional ultrasound images of each thyroid nodule and made a diagnosis of “benign” or “malignant” based on the TI-RADS category. After referring to S-Detect results, they re-evaluated the diagnoses. The diagnostic performance of radiologists was analyzed before and after referring to the results of S-Detect.Results: The accuracy, sensitivity, specificity, positive predictive value and negative predictive value of S-Detect were 77.0, 91.3, 65.2, 68.3 and 90.1%, respectively. In comparison with the less experienced radiologists (radiologist 1 and 2), S-Detect had a higher area under receiver operating characteristic curve (AUC), accuracy and specificity (p <0.05). In comparison with the most experienced radiologist, the diagnostic accuracy and AUC were lower (p<0.05). In the less experienced radiologists, the diagnostic accuracy, specificity and AUC were significantly improved when combined with S-Detect (p<0.05), but not for experienced radiologists (radiologist 3 and 4) (p>0.05).Conclusions: S-Detect may become an additional diagnostic method for the diagnosis of thyroid nodules and improve the diagnostic performance of less experienced radiologists. 

scholarly journals Comparisons of deep learning algorithms for diagnosing bacterial keratitis via external eye photographs

2021 ◽  
Vol 11 (1) ◽  
Author(s):  
Ming-Tse Kuo ◽  
Benny Wei-Yun Hsu ◽  
Yi-Sheng Lin ◽  
Po-Chiung Fang ◽  
Hun-Ju Yu ◽  
...  
Keyword(s):  
Deep Learning ◽  
Diagnostic Accuracy ◽  
Rural Areas ◽  
Predictive Value ◽  
Rapid Development ◽  
Characteristic Curve ◽  
Bacterial Keratitis ◽  
Average Percentage ◽  
Sensitivity Specificity ◽  
Corneal Melting

AbstractBacterial keratitis (BK), a painful and fulminant bacterial infection of the cornea, is the most common type of vision-threatening infectious keratitis (IK). A rapid clinical diagnosis by an ophthalmologist may often help prevent BK patients from progression to corneal melting or even perforation, but many rural areas cannot afford an ophthalmologist. Thanks to the rapid development of deep learning (DL) algorithms, artificial intelligence via image could provide an immediate screening and recommendation for patients with red and painful eyes. Therefore, this study aims to elucidate the potentials of different DL algorithms for diagnosing BK via external eye photos. External eye photos of clinically suspected IK were consecutively collected from five referral centers. The candidate DL frameworks, including ResNet50, ResNeXt50, DenseNet121, SE-ResNet50, EfficientNets B0, B1, B2, and B3, were trained to recognize BK from the photo toward the target with the greatest area under the receiver operating characteristic curve (AUROC). Via five-cross validation, EfficientNet B3 showed the most excellent average AUROC, in which the average percentage of sensitivity, specificity, positive predictive value, and negative predictive value was 74, 64, 77, and 61. There was no statistical difference in diagnostic accuracy and AUROC between any two of these DL frameworks. The diagnostic accuracy of these models (ranged from 69 to 72%) is comparable to that of the ophthalmologist (66% to 74%). Therefore, all these models are promising tools for diagnosing BK in first-line medical care units without ophthalmologists.

scholarly journals Deep learning and ensemble stacking technique for differentiating polypoidal choroidal vasculopathy from neovascular age-related macular degeneration

2021 ◽  
Vol 11 (1) ◽  
Author(s):  
Yu-Bai Chou ◽  
Chung-Hsuan Hsu ◽  
Wei-Shiang Chen ◽  
Shih-Jen Chen ◽  
De-Kuang Hwang ◽  
...  
Keyword(s):  
Deep Learning ◽  
Macular Degeneration ◽  
Polypoidal Choroidal Vasculopathy ◽  
Multiple Correspondence Analysis ◽  
Characteristic Curve ◽  
Medical Engineering ◽  
Age Related Macular Degeneration ◽  
Age Related ◽  
Sensitivity Specificity ◽  
Choroidal Vasculopathy

AbstractPolypoidal choroidal vasculopathy (PCV) and neovascular age-related macular degeneration (nAMD) share some similarity in clinical imaging manifestations. However, their disease entity and treatment strategy as well as visual outcomes are very different. To distinguish these two vision-threatening diseases is somewhat challenging but necessary. In this study, we propose a new artificial intelligence model using an ensemble stacking technique, which combines a color fundus photograph-based deep learning (DL) model and optical coherence tomography-based biomarkers, for differentiation of PCV from nAMD. Furthermore, we introduced multiple correspondence analysis, a method of transforming categorical data into principal components, to handle the dichotomous data for combining with another image DL system. This model achieved a robust performance with an accuracy, sensitivity, specificity, and area under the receiver operating characteristic curve of 83.67%, 80.76%, 84.72%, and 88.57%, respectively, by training nearly 700 active cases with suitable imaging quality and transfer learning architecture. This work could offer an alternative method of developing a multimodal DL model, improve its efficiency for distinguishing different diseases, and facilitate the broad application of medical engineering in a DL model design.

scholarly journals The Clinical Influence after Implementation of Convolutional Neural Network-Based Software for Diabetic Retinopathy Detection in the Primary Care Setting

Life ◽  
2021 ◽  
Vol 11 (3) ◽  
pp. 200
Author(s):  
Yu-Hsuan Li ◽  
Wayne Huey-Herng Sheu ◽  
Chien-Chih Chou ◽  
Chun-Hsien Lin ◽  
Yuan-Shao Cheng ◽  
...  
Keyword(s):  
Diabetic Retinopathy ◽  
Deep Learning ◽  
Operating Characteristic ◽  
Care Setting ◽  
Characteristic Curve ◽  
Clinical Implementation ◽  
Clinical Workflow ◽  
Wide Range ◽  
Before And After ◽  
Sensitivity Specificity

Deep learning-based software is developed to assist physicians in terms of diagnosis; however, its clinical application is still under investigation. We integrated deep-learning-based software for diabetic retinopathy (DR) grading into the clinical workflow of an endocrinology department where endocrinologists grade for retinal images and evaluated the influence of its implementation. A total of 1432 images from 716 patients and 1400 images from 700 patients were collected before and after implementation, respectively. Using the grading by ophthalmologists as the reference standard, the sensitivity, specificity, and area under the receiver operating characteristic curve (AUC) to detect referable DR (RDR) were 0.91 (0.87–0.96), 0.90 (0.87–0.92), and 0.90 (0.87–0.93) at the image level; and 0.91 (0.81–0.97), 0.84 (0.80–0.87), and 0.87 (0.83–0.91) at the patient level. The monthly RDR rate dropped from 55.1% to 43.0% after implementation. The monthly percentage of finishing grading within the allotted time increased from 66.8% to 77.6%. There was a wide range of agreement values between the software and endocrinologists after implementation (kappa values of 0.17–0.65). In conclusion, we observed the clinical influence of deep-learning-based software on graders without the retinal subspecialty. However, the validation using images from local datasets is recommended before clinical implementation.

Sign in / Sign up

Export Citation Format

Share Document