scholarly journals Deep learning systems detect dysplasia with human-like accuracy using histopathology and probe-based confocal laser endomicroscopy

2021 ◽  
Vol 11 (1) ◽  
Author(s):  
Shan Guleria ◽  
Tilak U. Shah ◽  
J. Vincent Pulido ◽  
Matthew Fasullo ◽  
Lubaina Ehsan ◽  
...  
Keyword(s):  
Deep Learning ◽  
Diagnostic Accuracy ◽  
High Sensitivity ◽  
Confocal Laser Endomicroscopy ◽  
Confocal Laser ◽  
Learning Approaches ◽  
Learning Models ◽  
Whole Slide Image ◽  
Slide Image ◽  
Level Model

AbstractProbe-based confocal laser endomicroscopy (pCLE) allows for real-time diagnosis of dysplasia and cancer in Barrett’s esophagus (BE) but is limited by low sensitivity. Even the gold standard of histopathology is hindered by poor agreement between pathologists. We deployed deep-learning-based image and video analysis in order to improve diagnostic accuracy of pCLE videos and biopsy images. Blinded experts categorized biopsies and pCLE videos as squamous, non-dysplastic BE, or dysplasia/cancer, and deep learning models were trained to classify the data into these three categories. Biopsy classification was conducted using two distinct approaches—a patch-level model and a whole-slide-image-level model. Gradient-weighted class activation maps (Grad-CAMs) were extracted from pCLE and biopsy models in order to determine tissue structures deemed relevant by the models. 1970 pCLE videos, 897,931 biopsy patches, and 387 whole-slide images were used to train, test, and validate the models. In pCLE analysis, models achieved a high sensitivity for dysplasia (71%) and an overall accuracy of 90% for all classes. For biopsies at the patch level, the model achieved a sensitivity of 72% for dysplasia and an overall accuracy of 90%. The whole-slide-image-level model achieved a sensitivity of 90% for dysplasia and 94% overall accuracy. Grad-CAMs for all models showed activation in medically relevant tissue regions. Our deep learning models achieved high diagnostic accuracy for both pCLE-based and histopathologic diagnosis of esophageal dysplasia and its precursors, similar to human accuracy in prior studies. These machine learning approaches may improve accuracy and efficiency of current screening protocols.

scholarly journals Diagnostic Accuracy of Confocal Laser Endomicroscopy for the Diagnosis of Oral Squamous Cell Carcinoma: A Systematic Review and Meta-Analysis

2021 ◽  
Vol 18 (23) ◽  
pp. 12390
Author(s):  
Sneha Sethi ◽  
Xiangqun Ju ◽  
Richard M. Logan ◽  
Paul Sambrook ◽  
Robert A. McLaughlin ◽  
...  
Keyword(s):  
Systematic Review ◽  
Squamous Cell Carcinoma ◽  
Diagnostic Accuracy ◽  
Sensitivity And Specificity ◽  
Meta Analysis ◽  
High Sensitivity ◽  
Risk Of Bias ◽  
Confocal Laser Endomicroscopy ◽  
Confocal Laser ◽  
Likelihood Ratios

Background: Advances in treatment approaches for patients with oral squamous cell carcinoma (OSCC) have been unsuccessful in preventing frequent recurrences and distant metastases, leading to a poor prognosis. Early detection and prevention enable an improved 5-year survival and better prognosis. Confocal Laser Endomicroscopy (CLE) is a non-invasive imaging instrument that could enable an earlier diagnosis and possibly help in reducing unnecessary invasive surgical procedures. Objective: To present an up to date systematic review and meta-analysis assessing the diagnostic accuracy of CLE in diagnosing OSCC. Materials and Methods. PubMed, Scopus, and Web of Science databases were explored up to 30 June 2021, to collect articles concerning the diagnosis of OSCC through CLE. Screening: data extraction and appraisal was done by two reviewers. The quality of the methodology followed by the studies included in this review was assessed using the Quality Assessment of Diagnostic Accuracy Studies-2 (QUADAS-2) tool. A random effects model was used for the meta-analysis. Results: Six studies were included, leading to a total number of 361 lesions in 213 patients. The pooled sensitivity and specificity were 95% (95% CI, 92–97%; I2 = 77.5%) and 93% (95% CI, 90–95%; I2 = 68.6%); the pooled positive likelihood ratios and negative likelihood ratios were 10.85 (95% CI, 5.4–21.7; I2 = 55.9%) and 0.08 (95% CI, 0.03–0.2; I2 = 83.5%); and the pooled diagnostic odds ratio was 174.45 (95% CI, 34.51–881.69; I2 = 73.6%). Although risk of bias and heterogeneity is observed, this study validates that CLE may have a noteworthy clinical influence on the diagnosis of OSCC, through its high sensitivity and specificity. Conclusions: This review indicates an exceptionally high sensitivity and specificity of CLE for diagnosing OSCC. Whilst it is a promising diagnostic instrument, the limited number of existing studies and potential risk of bias of included studies does not allow us to draw firm conclusions. A conclusive inference can be drawn when more studies, possibly with homogeneous methodological approach, are performed.

scholarly journals Vulnerability in Deep Transfer Learning Models to Adversarial Fast Gradient Sign Attack for COVID-19 Prediction from Chest Radiography Images

2021 ◽  
Vol 11 (9) ◽  
pp. 4233
Author(s):  
Biprodip Pal ◽  
Debashis Gupta ◽  
Md. Rashed-Al-Mahfuz ◽  
Salem A. Alyami ◽  
Mohammad Ali Moni
Keyword(s):  
Deep Learning ◽  
Transfer Learning ◽  
Chest Radiography ◽  
High Sensitivity ◽  
Ct Images ◽  
Classification Performance ◽  
Learning Models ◽  
X Ray ◽  
Image Capturing ◽  
Fast Gradient

The COVID-19 pandemic requires the rapid isolation of infected patients. Thus, high-sensitivity radiology images could be a key technique to diagnose patients besides the polymerase chain reaction approach. Deep learning algorithms are proposed in several studies to detect COVID-19 symptoms due to the success in chest radiography image classification, cost efficiency, lack of expert radiologists, and the need for faster processing in the pandemic area. Most of the promising algorithms proposed in different studies are based on pre-trained deep learning models. Such open-source models and lack of variation in the radiology image-capturing environment make the diagnosis system vulnerable to adversarial attacks such as fast gradient sign method (FGSM) attack. This study therefore explored the potential vulnerability of pre-trained convolutional neural network algorithms to the FGSM attack in terms of two frequently used models, VGG16 and Inception-v3. Firstly, we developed two transfer learning models for X-ray and CT image-based COVID-19 classification and analyzed the performance extensively in terms of accuracy, precision, recall, and AUC. Secondly, our study illustrates that misclassification can occur with a very minor perturbation magnitude, such as 0.009 and 0.003 for the FGSM attack in these models for X-ray and CT images, respectively, without any effect on the visual perceptibility of the perturbation. In addition, we demonstrated that successful FGSM attack can decrease the classification performance to 16.67% and 55.56% for X-ray images, as well as 36% and 40% in the case of CT images for VGG16 and Inception-v3, respectively, without any human-recognizable perturbation effects in the adversarial images. Finally, we analyzed that correct class probability of any test image which is supposed to be 1, can drop for both considered models and with increased perturbation; it can drop to 0.24 and 0.17 for the VGG16 model in cases of X-ray and CT images, respectively. Thus, despite the need for data sharing and automated diagnosis, practical deployment of such program requires more robustness.

scholarly journals Deep Learning with Neuroimaging and Genomics in Alzheimer’s Disease

2021 ◽  
Vol 22 (15) ◽  
pp. 7911
Author(s):  
Eugene Lin ◽  
Chieh-Hsin Lin ◽  
Hsien-Yuan Lane
Keyword(s):  
Alzheimer’S Disease ◽  
Alzheimer's Disease ◽  
Deep Learning ◽  
Future Research ◽  
Learning Approaches ◽  
Learning Models ◽  
Learning Techniques ◽  
Neuroimaging Data ◽  
Similarities And Differences ◽  
Normal Controls

A growing body of evidence currently proposes that deep learning approaches can serve as an essential cornerstone for the diagnosis and prediction of Alzheimer’s disease (AD). In light of the latest advancements in neuroimaging and genomics, numerous deep learning models are being exploited to distinguish AD from normal controls and/or to distinguish AD from mild cognitive impairment in recent research studies. In this review, we focus on the latest developments for AD prediction using deep learning techniques in cooperation with the principles of neuroimaging and genomics. First, we narrate various investigations that make use of deep learning algorithms to establish AD prediction using genomics or neuroimaging data. Particularly, we delineate relevant integrative neuroimaging genomics investigations that leverage deep learning methods to forecast AD on the basis of incorporating both neuroimaging and genomics data. Moreover, we outline the limitations as regards to the recent AD investigations of deep learning with neuroimaging and genomics. Finally, we depict a discussion of challenges and directions for future research. The main novelty of this work is that we summarize the major points of these investigations and scrutinize the similarities and differences among these investigations.

scholarly journals Probe-based confocal laser endomicroscopy and Barrett's esophagus – just a scientific toy or significant improvement in diagnosis?

2021 ◽  
Author(s):  
Luka Vranić ◽  
Tin Nadarević ◽  
Davor Štimac
Keyword(s):  
Diagnostic Accuracy ◽  
Real Time ◽  
Barrett’S Esophagus ◽  
Predictive Value ◽  
Barrett's Esophagus ◽  
High Reliability ◽  
Confocal Laser Endomicroscopy ◽  
Confocal Laser ◽  
Imaging Method ◽  
Increased Risk

Background: Barrett’s esophagus (BE) requires surveillance to identify potential neoplasia at early stage. Standard surveillance regimen includes random four-quadrant biopsies by Seattle protocol. Main limitations of random biopsies are high risk of sampling error, difficulties in histology interpretation, common inadequate classification of pathohistological changes, increased risk of bleeding and time necessary to acquire the final diagnosis. Probe-based confocal laser endomicroscopy (pCLE) has emerged as a potential tool with an aim to overcome these obvious limitations. Summary: pCLE represents real-time microscopic imaging method that offers evaluation of epithelial and subepithelial structures with 1000-fold magnification. In theory, pCLE has potential to eliminate the need for biopsy in BE patient. The main advantages would be real-time diagnosis and decision making, greater diagnostic accuracy and to evaluate larger area compared to random biopsies. Clinical pCLE studies in esophagus show high diagnostic accuracy and its high negative predictive value offers high reliability and confidence to exclude dysplastic and neoplastic lesions. However, it still cannot replace histopathology due to lower positive predictive value and sensitivity. Key messages: Despite promising results, its role in routine use in patients with Barrett’s esophagus remains questionable primarily due to lack of well-organized double-blind randomized trials.

scholarly journals PlotMI: visualization of pairwise interactions and positional preferences learned by a deep learning model from sequence data

2021 ◽  
Author(s):  
Tuomo Hartonen ◽  
Teemu Kivioja ◽  
Jussi Taipale
Keyword(s):  
Machine Learning ◽  
Deep Learning ◽  
Sequence Data ◽  
Predictive Performance ◽  
Learning Model ◽  
Biological Research ◽  
Learning Approaches ◽  
Learning Models ◽  
Model Interpretation ◽  
Pairwise Interactions

Deep learning models have in recent years gained success in various tasks related to understanding information coded in the DNA sequence. Rapidly developing genome-wide measurement technologies provide large quantities of data ideally suited for modeling using deep learning or other powerful machine learning approaches. Although offering state-of-the art predictive performance, the predictions made by deep learning models can be difficult to understand. In virtually all biological research, the understanding of how a predictive model works is as important as the raw predictive performance. Thus interpretation of deep learning models is an emerging hot topic especially in context of biological research. Here we describe plotMI, a mutual information based model interpretation strategy that can intuitively visualize positional preferences and pairwise interactions learned by any machine learning model trained on sequence data with a defined alphabet as input. PlotMI is freely available at https://github.com/hartonen/plotMI.

scholarly journals Covid-19 detection via deep neural network and occlusion sensitivity maps

2021 ◽  
Author(s):  
Noor Ahmad ◽  
Muhammad Aminu ◽  
Mohd Halim Mohd Noor
Keyword(s):  
Neural Network ◽  
Deep Learning ◽  
Deep Neural Network ◽  
State Of The Art ◽  
Color Images ◽  
Fine Tuning ◽  
Training Dataset ◽  
Learning Approaches ◽  
Learning Models ◽  
Sensitivity Maps

Deep learning approaches have attracted a lot of attention in the automatic detection of Covid-19 and transfer learning is the most common approach. However, majority of the pre-trained models are trained on color images, which can cause inefficiencies when fine-tuning the models on Covid-19 images which are often grayscale. To address this issue, we propose a deep learning architecture called CovidNet which requires a relatively smaller number of parameters. CovidNet accepts grayscale images as inputs and is suitable for training with limited training dataset. Experimental results show that CovidNet outperforms other state-of-the-art deep learning models for Covid-19 detection.

scholarly journals Deep Learning Algorithm for the Confirmation of Mucosal Healing in Crohn’s Disease, Based on Confocal Laser Endomicroscopy Images

2021 ◽  
Vol 30 (1) ◽  
pp. 59-65
Author(s):  
Anca Loredana Udristoiu ◽  
Daniela Stefanescu ◽  
Gabriel Gruionu ◽  
Lucian Gheorghe Gruionu ◽  
Andreea Valentina Iacob ◽  
...  
Keyword(s):  
Crohn’S Disease ◽  
Deep Learning ◽  
Crohn's Disease ◽  
Colonic Mucosa ◽  
Mucosal Healing ◽  
Machine Learning Algorithms ◽  
Confocal Laser Endomicroscopy ◽  
Confocal Laser ◽  
Normal Colonic Mucosa ◽  
Test Accuracy

Background and Aims: Mucosal healing (MH) is associated with a stable course of Crohn’s disease (CD) which can be assessed by confocal laser endomicroscopy (CLE). To minimize the operator’s errors and automate assessment of CLE images, we used a deep learning (DL) model for image analysis. We hypothesized that DL combined with convolutional neural networks (CNNs) and long short-term memory (LSTM) can distinguish between normal and inflamed colonic mucosa from CLE images. Methods: The study included 54 patients, 32 with known active CD, and 22 control patients (18 CD patients with MH and four normal mucosa patients with no history of inflammatory bowel diseases). We designed and trained a deep convolutional neural network to detect active CD using 6,205 endomicroscopy images classified as active CD inflammation (3,672 images) and control mucosal healing or no inflammation (2,533 images). CLE imaging was performed on four colorectal areas and the terminal ileum. Gold standard was represented by the histopathological evaluation. The dataset was randomly split in two distinct training and testing datasets: 80% data from each patient were used for training and the remaining 20% for testing. The training dataset consists of 2,892 images with inflammation and 2,189 control images. The testing dataset consists of 780 images with inflammation and 344 control images of the colon. We used a CNN-LSTM model with four convolution layers and one LSTM layer for automatic detection of MH and CD diagnosis from CLE images. Results: CLE investigation reveals normal colonic mucosa with round crypts and inflamed mucosa with irregular crypts and tortuous and dilated blood vessels. Our method obtained a 95.3% test accuracy with a specificity of 92.78% and a sensitivity of 94.6%, with an area under each receiver operating characteristic curves of 0.98. Conclusions: Using machine learning algorithms on CLE images can successfully differentiate between inflammation and normal ileocolonic mucosa and can be used as a computer aided diagnosis for CD. Future clinical studies with a larger patient spectrum will validate our results and improve the CNN-SSTM model.

Sign in / Sign up

Export Citation Format

Share Document