Dataset Information

Real-time computer-aided diagnosis of focal pancreatic masses from endoscopic ultrasound imaging based on a hybrid convolutional and long short-term memory neural network model.

ABSTRACT: Differential diagnosis of focal pancreatic masses is based on endoscopic ultrasound (EUS) guided fine needle aspiration biopsy (EUS-FNA/FNB). Several imaging techniques (i.e. gray-scale, color Doppler, contrast-enhancement and elastography) are used for differential diagnosis. However, diagnosis remains highly operator dependent. To address this problem, machine learning algorithms (MLA) can generate an automatic computer-aided diagnosis (CAD) by analyzing a large number of clinical images in real-time. We aimed to develop a MLA to characterize focal pancreatic masses during the EUS procedure. The study included 65 patients with focal pancreatic masses, with 20 EUS images selected from each patient (grayscale, color Doppler, arterial and venous phase contrast-enhancement and elastography). Images were classified based on cytopathology exam as: chronic pseudotumoral pancreatitis (CPP), neuroendocrine tumor (PNET) and ductal adenocarcinoma (PDAC). The MLA is based on a deep learning method which combines convolutional (CNN) and long short-term memory (LSTM) neural networks. 2688 images were used for training and 672 images for testing the deep learning models. The CNN was developed to identify the discriminative features of images, while a LSTM neural network was used to extract the dependencies between images. The model predicted the clinical diagnosis with an area under curve index of 0.98 and an overall accuracy of 98.26%. The negative (NPV) and positive (PPV) predictive values and the corresponding 95% confidential intervals (CI) are 96.7%, [94.5, 98.9] and 98.1%, [96.81, 99.4] for PDAC, 96.5%, [94.1, 98.8], and 99.7%, [99.3, 100] for CPP, and 98.9%, [97.5, 100] and 98.3%, [97.1, 99.4] for PNET. Following further validation on a independent test cohort, this method could become an efficient CAD tool to differentiate focal pancreatic masses in real-time.

SUBMITTER: Udristoiu AL

PROVIDER: S-EPMC8238220 | biostudies-literature |

REPOSITORIES: biostudies-literature

ACCESS DATA

Similar Datasets

Project description:Training deep Convolutional Neural Networks (CNNs) presents challenges in terms of memory requirements and computational resources, often resulting in issues such as model overfitting and lack of generalization. These challenges can only be mitigated by using an excessive number of training images. However, medical image datasets commonly suffer from data scarcity due to the complexities involved in their acquisition, preparation, and curation. To address this issue, we propose a compact and hybrid machine learning architecture based on the Morphological and Convolutional Neural Network (MCNN), followed by a Random Forest classifier. Unlike deep CNN architectures, the MCNN was specifically designed to achieve effective performance with medical image datasets limited to a few hundred samples. It incorporates various morphological operations into a single layer and uses independent neural networks to extract information from each signal channel. The final classification is obtained by utilizing a Random Forest classifier on the outputs of the last neural network layer. We compare the classification performance of our proposed method with three popular deep CNN architectures (ResNet-18, ShuffleNet-V2, and MobileNet-V2) using two training approaches: full training and transfer learning. The evaluation was conducted on two distinct medical image datasets: the ISIC dataset for melanoma classification and the ORIGA dataset for glaucoma classification. Results demonstrate that the MCNN method exhibits reliable performance in melanoma classification, achieving an AUC of 0.94 (95% CI: 0.91 to 0.97), outperforming the popular CNN architectures. For the glaucoma dataset, the MCNN achieved an AUC of 0.65 (95% CI: 0.53 to 0.74), which was similar to the performance of the popular CNN architectures. This study contributes to the understanding of mathematical morphology in shallow neural networks for medical image classification and highlights the potential of hybrid architectures in effectively learning from medical image datasets that are limited by a small number of case samples.

Project description:To improve the performance of a computer-aided detection (CAD) system for mass detection by using four-view information in screening mammography.The authors developed a four-view CAD system that emulates radiologists' reading by using the craniocaudal and mediolateral oblique views of the ipsilateral breast to reduce false positives (FPs) and the corresponding views of the contralateral breast to detect asymmetry. The CAD system consists of four major components: (1) Initial detection of breast masses on individual views, (2) information fusion of the ipsilateral views of the breast (referred to as two-view analysis), (3) information fusion of the corresponding views of the contralateral breast (referred to as bilateral analysis), and (4) fusion of the four-view information with a decision tree. The authors collected two data sets for training and testing of the CAD system: A mass set containing 389 patients with 389 biopsy-proven masses and a normal set containing 200 normal subjects. All cases had four-view mammograms. The true locations of the masses on the mammograms were identified by an experienced MQSA radiologist. The authors randomly divided the mass set into two independent sets for cross validation training and testing. The overall test performance was assessed by averaging the free response receiver operating characteristic (FROC) curves of the two test subsets. The FP rates during the FROC analysis were estimated by using the normal set only. The jackknife free-response ROC (JAFROC) method was used to estimate the statistical significance of the difference between the test FROC curves obtained with the single-view and the four-view CAD systems.Using the single-view CAD system, the breast-based test sensitivities were 58% and 77% at the FP rates of 0.5 and 1.0 per image, respectively. With the four-view CAD system, the breast-based test sensitivities were improved to 76% and 87% at the corresponding FP rates, respectively. The improvement was found to be statistically significant (p < 0.0001) by JAFROC analysis.The four-view information fusion approach that emulates radiologists' reading strategy significantly improves the performance of breast mass detection of the CAD system in comparison with the single-view approach.

Project description:Purpose To assess the performance of computer-aided diagnosis (CAD) systems and to determine the dominant ultrasonographic (US) features when classifying benign versus malignant focal liver lesions (FLLs) by using contrast material-enhanced US cine clips. Materials and Methods One hundred six US data sets in all subjects enrolled by three centers from a multicenter trial that included 54 malignant, 51 benign, and one indeterminate FLL were retrospectively analyzed. The 105 benign or malignant lesions were confirmed at histologic examination, contrast-enhanced computed tomography (CT), dynamic contrast-enhanced magnetic resonance (MR) imaging, and/or 6 or more months of clinical follow-up. Data sets included 3-minute cine clips that were automatically corrected for in-plane motion and automatically filtered out frames acquired off plane. B-mode and contrast-specific features were automatically extracted on a pixel-by-pixel basis and analyzed by using an artificial neural network (ANN) and a support vector machine (SVM). Areas under the receiver operating characteristic curve (AUCs) for CAD were compared with those for one experienced and one inexperienced blinded reader. A third observer graded cine quality to assess its effects on CAD performance. Results CAD, the inexperienced observer, and the experienced observer were able to analyze 95, 100, and 102 cine clips, respectively. The AUCs for the SVM, ANN, and experienced and inexperienced observers were 0.883 (95% confidence interval [CI]: 0.793, 0.940), 0.829 (95% CI: 0.724, 0.901), 0.843 (95% CI: 0.756, 0.903), and 0.702 (95% CI: 0.586, 0.782), respectively; only the difference between SVM and the inexperienced observer was statistically significant. Accuracy improved from 71.3% (67 of 94; 95% CI: 60.6%, 79.8%) to 87.7% (57 of 65; 95% CI: 78.5%, 93.8%) and from 80.9% (76 of 94; 95% CI: 72.3%, 88.3%) to 90.3% (65 of 72; 95% CI: 80.6%, 95.8%) when CAD was in agreement with the inexperienced reader and when it was in agreement with the experienced reader, respectively. B-mode heterogeneity and contrast material washout were the most discriminating features selected by CAD for all iterations. CAD selected time-based time-intensity curve (TIC) features 99.0% (207 of 209) of the time to classify FLLs, versus 1.0% (two of 209) of the time for intensity-based features. None of the 15 video-quality criteria had a statistically significant effect on CAD accuracy-all P values were greater than the Holm-Sidak α-level correction for multiple comparisons. Conclusion CAD systems classified benign and malignant FLLs with an accuracy similar to that of an expert reader. CAD improved the accuracy of both readers. Time-based features of TIC were more discriminating than intensity-based features. © RSNA, 2017 Online supplemental material is available for this article.

Project description:To evaluate the interobserver variability in descriptions of breast masses by dedicated breast imagers and radiology residents and determine how any differences in lesion description affect the performance of a computer-aided diagnosis (CAD) computer classification system.Institutional review board approval was obtained for this HIPAA-compliant study, and the requirement to obtain informed consent was waived. Images of 50 breast lesions were individually interpreted by seven dedicated breast imagers and 10 radiology residents, yielding 850 lesion interpretations. Lesions were described with use of 11 descriptors from the Breast Imaging Reporting and Data System, and interobserver variability was calculated with the Cohen κ statistic. Those 11 features were selected, along with patient age, and merged together by a linear discriminant analysis (LDA) classification model trained by using 1005 previously existing cases. Variability in the recommendations of the computer model for different observers was also calculated with the Cohen κ statistic.A significant difference was observed for six lesion features, and radiology residents had greater interobserver variability in their selection of five of the six features than did dedicated breast imagers. The LDA model accurately classified lesions for both sets of observers (area under the receiver operating characteristic curve = 0.94 for residents and 0.96 for dedicated imagers). Sensitivity was maintained at 100% for residents and improved from 98% to 100% for dedicated breast imagers. For residents, the computer model could potentially improve the specificity from 20% to 40% (P < .01) and the κ value from 0.09 to 0.53 (P < .001). For dedicated breast imagers, the computer model could increase the specificity from 34% to 43% (P = .16) and the κ value from 0.21 to 0.61 (P < .001).Among findings showing a significant difference, there was greater interobserver variability in lesion descriptions among residents; however, an LDA model using data from either dedicated breast imagers or residents yielded a consistently high performance in the differentiation of benign from malignant breast lesions, demonstrating potential for improving specificity and decreasing interobserver variability in biopsy recommendations.

Dataset Information

Real-time computer-aided diagnosis of focal pancreatic masses from endoscopic ultrasound imaging based on a hybrid convolutional and long short-term memory neural network model.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets