Training Ultrasound Image Classification Deep-Learning Algorithms for Pneumothorax Detection Using a Synthetic Tissue Phantom Apparatus

Boice, Emily N.; Torres, Sofia I. Hernandez; Knowlton, Zechariah J.; Berard, David; Gonzalez, Jose M.; Avital, Guy; Snider, Eric J.

doi:10.3390/jimaging8090249

Cited by 15 publications

(15 citation statements)

References 37 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…A critical task in combat casualty care for ultrasound is the extended Focused Assessment with Sonography for Trauma (eFAST) examination for detecting fluid in the lungs or abdomen. A similar phantom development methodology and subsequent training could be used for tasks such as this, as we have already shown for pneumothorax [ 45 ], to create a more complex training environment for training neural network models.…”

Section: Discussionmentioning

confidence: 99%

Using an Ultrasound Tissue Phantom Model for Hybrid Training of Deep Learning Models for Shrapnel Detection

2022

View full text Add to dashboard Cite

Tissue phantoms are important for medical research to reduce the use of animal or human tissue when testing or troubleshooting new devices or technology. Development of machine-learning detection tools that rely on large ultrasound imaging data sets can potentially be streamlined with high quality phantoms that closely mimic important features of biological tissue. Here, we demonstrate how an ultrasound-compliant tissue phantom comprised of multiple layers of gelatin to mimic bone, fat, and muscle tissue types can be used for machine-learning training. This tissue phantom has a heterogeneous composition to introduce tissue level complexity and subject variability in the tissue phantom. Various shrapnel types were inserted into the phantom for ultrasound imaging to supplement swine shrapnel image sets captured for applications such as deep learning algorithms. With a previously developed shrapnel detection algorithm, blind swine test image accuracy reached more than 95% accuracy when training was comprised of 75% tissue phantom images, with the rest being swine images. For comparison, a conventional MobileNetv2 deep learning model was trained with the same training image set and achieved over 90% accuracy in swine predictions. Overall, the tissue phantom demonstrated high performance for developing deep learning models for ultrasound image classification.

show abstract

Section: Discussionmentioning

confidence: 99%

Using an Ultrasound Tissue Phantom Model for Hybrid Training of Deep Learning Models for Shrapnel Detection

2022

View full text Add to dashboard Cite

show abstract

“…Third, extension of this shrapnel application to animal and human image sets will be pursued as will its use in other areas of the body. Image classification of shrapnel in the chest abdomen can be paired with ongoing research for the extended focused assessment with sonography in trauma (eFAST) exam, a widely used procedure for detecting free fluid in the chest or abdomen in emergency or military medicine applications [44][45][46].…”

Section: Discussionmentioning

confidence: 99%

Using Ultrasound Image Augmentation and Ensemble Predictions to Prevent Machine-Learning Model Overfitting

2023

Self Cite

View full text Add to dashboard Cite

Deep learning predictive models have the potential to simplify and automate medical imaging diagnostics by lowering the skill threshold for image interpretation. However, this requires predictive models that are generalized to handle subject variability as seen clinically. Here, we highlight methods to improve test accuracy of an image classifier model for shrapnel identification using tissue phantom image sets. Using a previously developed image classifier neural network—termed ShrapML—blind test accuracy was less than 70% and was variable depending on the training/test data setup, as determined by a leave one subject out (LOSO) holdout methodology. Introduction of affine transformations for image augmentation or MixUp methodologies to generate additional training sets improved model performance and overall accuracy improved to 75%. Further improvements were made by aggregating predictions across five LOSO holdouts. This was done by bagging confidences or predictions from all LOSOs or the top-3 LOSO confidence models for each image prediction. Top-3 LOSO confidence bagging performed best, with test accuracy improved to greater than 85% accuracy for two different blind tissue phantoms. This was confirmed by gradient-weighted class activation mapping to highlight that the image classifier was tracking shrapnel in the image sets. Overall, data augmentation and ensemble prediction approaches were suitable for creating more generalized predictive models for ultrasound image analysis, a critical step for real-time diagnostic deployment.

show abstract

“…For cropping, pixel coordinates of the upper corner were identified, as well as the length and width of the window of interest. To boost the number of US images, a 4-pixel rolling window was used for further cropping, yielding 108 image segments per M-mode image, following the process described previously ( Boice et al, 2022a ). The image segments were then resized to 512 × 512 × 3 using MATLAB R2022b batch image processing.…”

Section: Methodsmentioning

confidence: 99%

“…Positive PTX was simulated by imaging the rib model with an air gap between it and the lung model. Quality of the images was validated by training a classifier algorithm on phantom images which resulted in more than 90% accuracy on blind PTX positive and negative images captured in euthanized swine tissue ( Boice et al, 2022a ). However, this phantom lacked the complete anatomical rib structure needed for integration into an eFAST phantom.…”

Section: Introductionmentioning

confidence: 99%

An extended focused assessment with sonography in trauma ultrasound tissue-mimicking phantom for developing automated diagnostic technologies

Hernandez-Torres,

Bedolla,

Berard

et al. 2023

Front. Bioeng. Biotechnol.

Self Cite

View full text Add to dashboard Cite

Introduction: Medical imaging-based triage is critical for ensuring medical treatment is timely and prioritized. However, without proper image collection and interpretation, triage decisions can be hard to make. While automation approaches can enhance these triage applications, tissue phantoms must be developed to train and mature these novel technologies. Here, we have developed a tissue phantom modeling the ultrasound views imaged during the enhanced focused assessment with sonography in trauma exam (eFAST).Methods: The tissue phantom utilized synthetic clear ballistic gel with carveouts in the abdomen and rib cage corresponding to the various eFAST scan points. Various approaches were taken to simulate proper physiology without injuries present or to mimic pneumothorax, hemothorax, or abdominal hemorrhage at multiple locations in the torso. Multiple ultrasound imaging systems were used to acquire ultrasound scans with or without injury present and were used to train deep learning image classification predictive models.Results: Performance of the artificial intelligent (AI) models trained in this study achieved over 97% accuracy for each eFAST scan site. We used a previously trained AI model for pneumothorax which achieved 74% accuracy in blind predictions for images collected with the novel eFAST tissue phantom. Grad-CAM heat map overlays for the predictions identified that the AI models were tracking the area of interest for each scan point in the tissue phantom.Discussion: Overall, the eFAST tissue phantom ultrasound scans resembled human images and were successful in training AI models. Tissue phantoms are critical first steps in troubleshooting and developing medical imaging automation technologies for this application that can accelerate the widespread use of ultrasound imaging for emergency triage.

show abstract

Training Ultrasound Image Classification Deep-Learning Algorithms for Pneumothorax Detection Using a Synthetic Tissue Phantom Apparatus

Cited by 15 publications

References 37 publications

Using an Ultrasound Tissue Phantom Model for Hybrid Training of Deep Learning Models for Shrapnel Detection

Using an Ultrasound Tissue Phantom Model for Hybrid Training of Deep Learning Models for Shrapnel Detection

Using Ultrasound Image Augmentation and Ensemble Predictions to Prevent Machine-Learning Model Overfitting

An extended focused assessment with sonography in trauma ultrasound tissue-mimicking phantom for developing automated diagnostic technologies

Contact Info

Product

Resources

About