Deep Convolutional Neural Networks for Chest Diseases Detection

Abiyev, Rahib H.; Ma’aitah, Mohammad Khaleel Sallam

doi:10.1155/2018/4168538

Cited by 247 publications

(113 citation statements)

References 35 publications

Supporting

Mentioning

111

Contrasting

Unclassified

Order By: Relevance

“…Multiple uses have been described for artificial intelligence in medicine, and medical imaging is consistently noted as one of the greatest potential uses. To date, most examples of clinically useful DL image interpretation algorithms have focused on radiology‐based implementations such as interpretations of chest radiography, computed tomography (CT), and magnetic resonance imaging (MRI) . Ultrasound (US) examples of DL use for image analysis are relatively few and far between, with most being found in high‐end consultative imaging–type US machines …”

mentioning

confidence: 99%

Are All Deep Learning Architectures Alike for Point‐of‐Care Ultrasound?: Evidence From a Cardiac Image Classification Model Suggests Otherwise

Blaivas

2019

J of Ultrasound Medicine

View full text Add to dashboard Cite

Objectives-Little is known about optimal deep learning (DL) approaches for point-of-care ultrasound (POCUS) applications. We compared 6 popular DL architectures for POCUS cardiac image classification to determine whether an optimal DL architecture exists for future DL algorithm development in POCUS.Methods-We trained 6 convolutional neural networks (CNNs) with a range of complexities and ages (AlexNet, VGG-16, VGG-19, ResNet50, Den-seNet201, and Inception-v4). Each CNN was trained by using images of 5 typical POCUS cardiac views. Images were extracted from 225 publicly available deidentified POCUS cardiac videos. A total of 750,018 individual images were extracted, with 90% used for model training and 10% for cross-validation. The training time and accuracy achieved were tracked. A real-world test of the algorithms was performed on a set of 125 completely new cardiac images. Descriptive statistics, Pearson R values, and κ values were calculated for each CNN.Results-Accuracy ranged from 96% to 85.6% correct for the 6 CNNs. VGG-16, one of the oldest and simplest CNNs, performed best at 96% correct with 232 minutes to train (R = 0.97; κ = 0.95; P < .00001). The worst-performing CNN was the newer DenseNet201, with 85.6% accuracy and 429 minutes to train (R = 0.92; κ = 0.82; P < .00001).Conclusions-Six common image classification DL algorithms showed considerable variability in their accuracy and training time when trained and tested on identical data, suggesting that not all will perform optimally for POCUS DL applications. Contrary to well-established accuracies for CNNs, more modern and deeper algorithms yielded poorer results.

show abstract

mentioning

confidence: 99%

Are All Deep Learning Architectures Alike for Point‐of‐Care Ultrasound?: Evidence From a Cardiac Image Classification Model Suggests Otherwise

Blaivas

2019

J of Ultrasound Medicine

View full text Add to dashboard Cite

show abstract

“…Within medicine, deep learning algorithms have shown particular promise in the machine interpretation of diagnostic imaging techniques across various organ systems. For example, the application of deep learning techniques to the interpretation of chest x‐rays and computed tomography (CT) scans of the head and chest have all been shown to yield improved diagnostic accuracy when compared to radiologists 1‐4 . However, some of these studies and algorithms have come under justified criticism for inadequate validation in real world applications 5 …”

Section: Introductionmentioning

confidence: 99%

DIY AI, deep learning network development for automated image classification in a point‐of‐care ultrasound quality assurance program

2020

View full text Add to dashboard Cite

Presented at 2019 ACEP Scientific Assembly at #344. Funding and support: By JACEP Open policy, all authors are required to disclose any and all commercial, financial, and other relationships in any way related to the subject of this article as per ICMJE conflict of interest guidelines (see www.icmje.org). The authors have stated that no such relationships exist. AbstractBackground: Artificial intelligence (AI) is increasingly a part of daily life and offers great possibilities to enrich health care. Imaging applications of AI have been mostly developed by large, well-funded companies and currently are inaccessible to the comparatively small market of point-of-care ultrasound (POCUS) programs. Given this absence of commercial solutions, we sought to create and test a do-it-yourself (DIY) deep learning algorithm to classify ultrasound images to enhance the quality assurance work-flow for POCUS programs. Methods:We created a convolutional neural network using publicly available software tools and pre-existing convolutional neural network architecture. The convolutional neural network was subsequently trained using ultrasound images from seven ultrasound exam types: pelvis, heart, lung, abdomen, musculoskeletal, ocular, and central vascular access from 189 publicly available POCUS videos. Approximately 121,000 individual images were extracted from the videos, 80% were used for model training and 10% each for cross validation and testing. We then tested the algorithm for accuracy against a set of 160 randomly extracted ultrasound frames from ultrasound videos not previously used for training and that were performed on different ultrasound equipment. Three POCUS experts blindly categorized the 160 random images, and results were compared to the convolutional neural network algorithm. Descriptive statistics and Krippendorff alpha reliability estimates were calculated. Results:The cross validation of the convolutional neural network approached 99% for accuracy. The algorithm accurately classified 98% of the test ultrasound images.In the new POCUS program simulation phase, the algorithm accurately classified 70% of 160 new images for moderate correlation with the ground truth, = 0.64. The three blinded POCUS experts correctly classified 93%, 94%, and 98% of the images,

show abstract

“…Chest X-Ray(CXR) gives in overall orientation as an underlying symptomatic examination and is particularly valuable in the conclusion of pneumonia, malignant growth and COPD [4]. By analyzing CXR image, radiologists can diagnose many lung related diseases [5]. Skilled radiologists use CXR to recognize diseases, for example tuberculosis, pneumonia, interstitial lung malady, and cancer [6].…”

Section: Introductionmentioning

confidence: 99%

“…Classifying CXR irregularities is considered as a dreary task for radiologists [5]. Many studies have been conducted in the domain of Artificial Intelligence to aid radiologists in reading CXR Images.…”

Section: Introductionmentioning

confidence: 99%

U-Net Based Lung Image Segmentation for Lung Disease Detection

2019

IJRTE

View full text Add to dashboard Cite

Lung diseases are becoming a worldwide health problem. World Health Organization estimates that by 2030, lung diseases such as Chronic Obstructive Pulmonary Disease will be one of the leading cause of mortality. Accurate and timely detection of lung diseases may prevent further death. It is therefore vital that its early detection will lead to treatment and prevention of mortality among patients. However, the scarcity of expert or well-trained radiologists reading CXR images might delay the timely diagnosis of lung diseases especially in rural areas where the scarcity is felt. In order to aid radiologist in reading CXR images, a computer aided tool is proposed for faster and more accurate reading of CXR images. To prepare the image for processing, it need to be segmented to make it easier for the computer to understand. The goal of image segmentation in medical field is to extract the region of interest in the organ. This study is focused on developing a model that will segment the lung from CXR images. Using U-Net architecture based semantic segmentation, the researchers were able to develop and train a model using a set of 562 CXR images and lung mask images, 70 percent of the images were used for training and 30 percent for testing. The developed model achieved a final training accuracy of 97.55 percent and validation accuracy of 97.37 percent. Validation loss and training loss are also low which indicates that the model can segment lung from CXR images with minimal error. The developed model can then be used in classifying lung diseases by focusing on the segmented image rather than focusing on the entire CXR image.

show abstract

Deep Convolutional Neural Networks for Chest Diseases Detection

Cited by 247 publications

References 35 publications

Are All Deep Learning Architectures Alike for Point‐of‐Care Ultrasound?: Evidence From a Cardiac Image Classification Model Suggests Otherwise

Are All Deep Learning Architectures Alike for Point‐of‐Care Ultrasound?: Evidence From a Cardiac Image Classification Model Suggests Otherwise

DIY AI, deep learning network development for automated image classification in a point‐of‐care ultrasound quality assurance program

U-Net Based Lung Image Segmentation for Lung Disease Detection

Contact Info

Product

Resources

About