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

Abstract

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,

Making Artificial Intelligence Lemonade Out of Data Lemons

“…Prior examples of POCUS ultrasound image augmentations have included modification of both still and dynamic images. 37 In one example researchers increased low numbers of adult inferior vena cava videos in a study on ML algorithm design for assessment of patient fluid responsiveness when being resuscitated for sepsis in the emergency department or intensive care unit. 16 Video rotations and flips allowed the researchers to convert a 191 video dataset into 3820 videos used for algorithm training.…”

Visual Rounds Based on Multiorgan Point-of-Care Ultrasound in the ICU

“…Automatic and quantitative scoring system was developed to integrate and standardize the clinical assessment of the lung ( 72 , 73 ). Classification of benchmark ultrasound images is required for training a convolutional artificial network ( 74 ). The 5G-powered robot-assisted teleultrasound diagnostic system has already been applied in intensive care units ( 75 ), which has an advantage as easy operation, good feasibility, comparable diagnosis, and deserves to promote widely in clinical practice.…”

Advanced Ultrasound and Photoacoustic Imaging in Cardiology

“…Another research area where DL is making significant progress is in improving the quality of image acquisition using POCUS [ 105 ]. Blaivas et al [ 106 ] developed a DL-based model for image quality assurance for automatic image classification. They used a large dataset of 121,000 images extracted from US sequences and had an accuracy of 98%.…”