An Interpretation Architecture for Deep Learning Models with the Application of COVID-19 Diagnosis

The Coronavirus disease 2019 (COVID-19) has become one of the threats to the world. Computed tomography (CT) is an informative tool for the diagnosis of COVID-19 patients. Many deep learning approaches on CT images have been proposed and brought promising performance. However, due to the high complexity and non-transparency of deep models, the explanation of the diagnosis process is challenging, making it hard to evaluate whether such approaches are reliable. In this paper, we propose a visual interpretation architecture for the explanation of the deep learning models and apply the architecture in COVID-19 diagnosis. Our architecture designs a comprehensive interpretation about the deep model from different perspectives, including the training trends, diagnostic performance, learned features, feature extractors, the hidden layers, the support regions for diagnostic decision, and etc. With the interpretation architecture, researchers can make a comparison and explanation about the classification performance, gain insight into what the deep model learned from images, and obtain the supports for diagnostic decisions. Our deep model achieves the diagnostic result of 94.75%, 93.22%, 96.69%, 97.27%, and 91.88% in the criteria of accuracy, sensitivity, specificity, positive predictive value, and negative predictive value, which are 8.30%, 4.32%, 13.33%, 10.25%, and 6.19% higher than that of the compared traditional methods. The visualized features in 2-D and 3-D spaces provide the reasons for the superiority of our deep model. Our interpretation architecture would allow researchers to understand more about how and why deep models work, and can be used as interpretation solutions for any deep learning models based on convolutional neural network. It can also help deep learning methods to take a step forward in the clinical COVID-19 diagnosis field.

Download Full-text

Artificial intelligence–based detection of aortic stenosis from chest radiographs

European Heart Journal - Digital Health ◽

10.1093/ehjdh/ztab102 ◽

2021 ◽

Author(s):

Daiju Ueda ◽

Akira Yamamoto ◽

Shoichi Ehara ◽

Shinichi Iwata ◽

Koji Abo ◽

...

Keyword(s):

Artificial Intelligence ◽

Deep Learning ◽

Aortic Stenosis ◽

Predictive Value ◽

Cost Effective ◽

Chest Radiographs ◽

Validation Dataset ◽

Learning Models ◽

Test Dataset ◽

Sensitivity Specificity

Abstract Aims We aimed to develop models to detect aortic stenosis (AS) from chest radiographs—one of the most basic imaging tests—with artificial intelligence. Methods and Results We used 10433 retrospectively collected digital chest radiographs from 5638 patients to train, validate, and test three deep learning models. Chest radiographs were collected from patients who had also undergone echocardiography at a single institution between July 2016 and May 2019. These were labelled from the corresponding echocardiography assessments as AS-positive or AS-negative. The radiographs were separated on a patient basis into training (8327 images from 4512 patients, mean age 65 ± [SD] 15 years), validation (1041 images from 563 patients, mean age 65 ± 14 years), and test (1065 images from 563 patients, mean age 65 ± 14 years) datasets. The soft voting-based ensemble of the three developed models had the best overall performance for predicting AS with an AUC, sensitivity, specificity, accuracy, positive predictive value, and negative predictive value of 0.83 (95% CI 0.77–0.88), 0.78 (0.67–0.86), 0.71 (0.68–0.73), 0.71 (0.68–0.74), 0.18 (0.14–0.23), and 0.97 (0.96–0.98), respectively, in the validation dataset and 0.83 (0.78–0.88), 0.83 (0.74–0.90), 0.69 (0.66–0.72), 0.71 (0.68–0.73), 0.23 (0.19–0.28), and 0.97 (0.96–0.98), respectively, in the test dataset. Conclusion Deep learning models using chest radiographs have the potential to differentiate between radiographs of patients with and without AS. Lay summary We created AI models using deep learning to identify aortic stenosis from chest radiographs. Three AI models were developed and evaluated with 10433 retrospectively collected radiographs and labelled from echocardiography reports. The ensemble AI model could detect aortic stenosis in a test dataset with an AUC of 0.83 (95% CI 0.78–0.88). Since chest radiography is a cost effective and widely available imaging test, our model can provide an additive resource for the detection of aortic stenosis.

Download Full-text

A Study on the Auxiliary Diagnosis of Thyroid Disease Images Based on Multiple Dimensional Deep Learning Algorithms

Current Medical Imaging Formerly Current Medical Imaging Reviews ◽

10.2174/1573405615666190115155223 ◽

2020 ◽

Vol 16 (3) ◽

pp. 199-205

Author(s):

Yuejun Liu ◽

Yifei Xu ◽

Xiangzheng Meng ◽

Xuguang Wang ◽

Tianxu Bai

Keyword(s):

Deep Learning ◽

Learning Algorithms ◽

Region Of Interest ◽

Classification Performance ◽

Thyroid Diseases ◽

Great Success ◽

Learning Models ◽

Good Classification Performance ◽

Spect Images

Background: Medical imaging plays an important role in the diagnosis of thyroid diseases. In the field of machine learning, multiple dimensional deep learning algorithms are widely used in image classification and recognition, and have achieved great success. Objective: The method based on multiple dimensional deep learning is employed for the auxiliary diagnosis of thyroid diseases based on SPECT images. The performances of different deep learning models are evaluated and compared. Methods: Thyroid SPECT images are collected with three types, they are hyperthyroidism, normal and hypothyroidism. In the pre-processing, the region of interest of thyroid is segmented and the amount of data sample is expanded. Four CNN models, including CNN, Inception, VGG16 and RNN, are used to evaluate deep learning methods. Results: Deep learning based methods have good classification performance, the accuracy is 92.9%-96.2%, AUC is 97.8%-99.6%. VGG16 model has the best performance, the accuracy is 96.2% and AUC is 99.6%. Especially, the VGG16 model with a changing learning rate works best. Conclusion: The standard CNN, Inception, VGG16, and RNN four deep learning models are efficient for the classification of thyroid diseases with SPECT images. The accuracy of the assisted diagnostic method based on deep learning is higher than that of other methods reported in the literature.

Download Full-text

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

Scientific Reports ◽

10.1038/s41598-021-84510-4 ◽

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.

Download Full-text

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

Applied Sciences ◽

10.3390/app11094233 ◽

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.

Download Full-text

Deep Learning with Neuroimaging and Genomics in Alzheimer’s Disease

International Journal of Molecular Sciences ◽

10.3390/ijms22157911 ◽

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.

Download Full-text

Additional value of deep learning computed tomographic angiography-based fractional flow reserve in detecting coronary stenosis and predicting outcomes

Acta Radiologica ◽

10.1177/0284185120983977 ◽

2021 ◽

pp. 028418512098397

Author(s):

Yang Li ◽

Hong Qiu ◽

Zhihui Hou ◽

Jianfeng Zheng ◽

Jianan Li ◽

...

Keyword(s):

Deep Learning ◽

Unstable Angina ◽

Fractional Flow Reserve ◽

Predictive Value ◽

Computed Tomographic Angiography ◽

Fractional Flow ◽

Computed Tomographic ◽

Flow Reserve ◽

Sensitivity Specificity ◽

Tomographic Angiography

Background Deep learning (DL) has achieved great success in medical imaging and could be utilized for the non-invasive calculation of fractional flow reserve (FFR) from coronary computed tomographic angiography (CCTA) (CT-FFR). Purpose To examine the ability of a DL-based CT-FFR in detecting hemodynamic changes of stenosis. Material and Methods This study included 73 patients (85 vessels) who were suspected of coronary artery disease (CAD) and received CCTA followed by invasive FFR measurements within 90 days. The diagnostic accuracy, sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), and area under the receiver operating characteristics curve (AUC) were compared between CT-FFR and CCTA. Thirty-nine patients who received drug therapy instead of revascularization were followed for up to 31 months. Major adverse cardiac events (MACE), unstable angina, and rehospitalization were evaluated and compared between the study groups. Results At the patient level, CT-FFR achieved 90.4%, 93.6%, 88.1%, 85.3%, and 94.9% in accuracy, sensitivity, specificity, PPV, and NPV, respectively. At the vessel level, CT-FFR achieved 91.8%, 93.9%, 90.4%, 86.1%, and 95.9%, respectively. CT-FFR exceeded CCTA in these measurements at both levels. The vessel-level AUC for CT-FFR also outperformed that for CCTA (0.957 vs. 0.599, P < 0.0001). Patients with CT-FFR ≤0.8 had higher rates of rehospitalization (hazard ratio [HR] 4.51, 95% confidence interval [CI] 1.08–18.9) and MACE (HR 7.26, 95% CI 0.88–59.8), as well as a lower rate of unstable angina (HR 0.46, 95% CI 0.07–2.91). Conclusion CT-FFR is superior to conventional CCTA in differentiating functional myocardial ischemia. In addition, it has the potential to differentiate prognoses of patients with CAD.

Download Full-text

Validating Deep Neural Networks for Online Decoding of Motor Imagery Movements from EEG Signals

Sensors ◽

10.3390/s19010210 ◽

2019 ◽

Vol 19 (1) ◽

pp. 210 ◽

Cited By ~ 32

Author(s):

Zied Tayeb ◽

Juri Fedjaev ◽

Nejla Ghaboosi ◽

Christoph Richter ◽

Lukas Everding ◽

...

Keyword(s):

Neural Network ◽

Machine Learning ◽

Deep Learning ◽

Convolutional Neural Network ◽

Motor Imagery ◽

Classification Performance ◽

Feature Engineering ◽

Learning Models ◽

Eeg Signals ◽

Learning Methods

Non-invasive, electroencephalography (EEG)-based brain-computer interfaces (BCIs) on motor imagery movements translate the subject’s motor intention into control signals through classifying the EEG patterns caused by different imagination tasks, e.g., hand movements. This type of BCI has been widely studied and used as an alternative mode of communication and environmental control for disabled patients, such as those suffering from a brainstem stroke or a spinal cord injury (SCI). Notwithstanding the success of traditional machine learning methods in classifying EEG signals, these methods still rely on hand-crafted features. The extraction of such features is a difficult task due to the high non-stationarity of EEG signals, which is a major cause by the stagnating progress in classification performance. Remarkable advances in deep learning methods allow end-to-end learning without any feature engineering, which could benefit BCI motor imagery applications. We developed three deep learning models: (1) A long short-term memory (LSTM); (2) a spectrogram-based convolutional neural network model (CNN); and (3) a recurrent convolutional neural network (RCNN), for decoding motor imagery movements directly from raw EEG signals without (any manual) feature engineering. Results were evaluated on our own publicly available, EEG data collected from 20 subjects and on an existing dataset known as 2b EEG dataset from “BCI Competition IV”. Overall, better classification performance was achieved with deep learning models compared to state-of-the art machine learning techniques, which could chart a route ahead for developing new robust techniques for EEG signal decoding. We underpin this point by demonstrating the successful real-time control of a robotic arm using our CNN based BCI.

Download Full-text

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

10.1101/2021.03.14.435285 ◽

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.

Download Full-text

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

10.36227/techrxiv.14100890 ◽

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.

Download Full-text

A Mask-guided Attention Deep Learning Model for COVID-19 Diagnosis based on an Integrated CT Scan Images Database

10.36227/techrxiv.18166667.v1 ◽

2022 ◽

Author(s):

Maede Maftouni ◽

Bo Shen ◽

Andrew Chung Chee Law ◽

Niloofar Ayoobi Yazdi ◽

Zhenyu Kong

Keyword(s):

Deep Learning ◽

Ct Scan ◽

Imaging Modality ◽

Learning Model ◽

Classification Performance ◽

Computer Assisted ◽

Learning Approach ◽

Learning Models ◽

Task Learning ◽

Data Efficiency

The global extent of COVID-19 mutations and the consequent depletion of hospital resources highlighted the necessity of effective computer-assisted medical diagnosis. COVID-19 detection mediated by deep learning models can help diagnose this highly contagious disease and lower infectivity and mortality rates. Computed tomography (CT) is the preferred imaging modality for building automatic COVID-19 screening and diagnosis models. It is well-known that the training set size significantly impacts the performance and generalization of deep learning models. However, accessing a large dataset of CT scan images from an emerging disease like COVID-19 is challenging. Therefore, data efficiency becomes a significant factor in choosing a learning model. To this end, we present a multi-task learning approach, namely, a mask-guided attention (MGA) classifier, to improve the generalization and data efficiency of COVID-19 classification on lung CT scan images.The novelty of this method is compensating for the scarcity of data by employing more supervision with lesion masks, increasing the sensitivity of the model to COVID-19 manifestations, and helping both generalization and classification performance. Our proposed model achieves better overall performance than the single-task baseline and state-of-the-art models, as measured by various popular metrics. In our experiment with different percentages of data from our curated dataset, the classification performance gain from this multi-task learning approach is more significant for the smaller training sizes. Furthermore, experimental results demonstrate that our method enhances the focus on the lesions, as witnessed by bothattention and attribution maps, resulting in a more interpretable model.

Download Full-text