Sort:
Open Access Original Research Issue
Evaluation of Liver Fibrosis on Grayscale Ultrasound in a Pediatric Population Using a Cloud-based Transfer Learning Artificial Intelligence Platform
Advanced Ultrasound in Diagnosis and Therapy 2024, 8(4): 242-249
Published: 30 December 2024
Abstract PDF (2.2 MB) Collect
Downloads:9
Objectives

The incidence of chronic liver diseases in children is increasing worldwide due to congenital, metabolic, autoimmune and viral diseases. Currently, liver biopsy for fibrosis assessment is considered the gold standard. However, this procedure is invasive, may result in unavoidable complications and is prone to sampling errors. These limitations have led to an increasing demand for noninvasive methods for fibrosis screening. Artificial intelligence integration in ultrasound diagnosis of liver fibrosis has gained interest in clinical research. In the current study we used a cloud-based artificial intelligence platform utilizing transfer learning to evaluate the accuracy of B-mode ultrasound based AI model compared to pediatric radiologists in detection of liver fibrosis in a pediatric population.

Methods

For this IRB approved study, charts of 190 pediatric patients who were referred for liver biopsy and ultrasound were reviewed. On average 14 images of different liver areas were selected and a single image per decision was used for both radiologist and AI reads. A supervised machine learning model for image classification was developed using Google Vision AutoML (Google Inc., Mountain View, CA, USA). Data was divided for model development (80% of cases (154 cases) = 2324 images) and a model validation cohort for external testing (20% (36 cases) = 360 images). As a comparator, three blinded radiologists read the ultrasound images of the validation cohort and provided a binary diagnosis of fibrosis versus non fibrotic liver appearance. Tissue sampling was used as the reference standard for all cases.

Results

There were 99 and 91 patients in the biopsy proven fibrosis and non-fibrosis group, respectively. The model’s internal evaluation resulted in precision of 78.2%, recall of 78.5% and average precision of 87.7%. In the external validation cohort, three radiologists (Mean ± Standard Deviation) and Google AutoML (confidence interval (CI)) achieved a sensitivity of 42.04% ± 0.04 and 70.56% (63.32% to 77.10% CI), specificity of 50.18% ± 0.04 and 45.00% (37.59% to 52.58% CI), positive predictive value of 45.76% ± 0.01 and 56.19% (52.17% to 60.14% CI), negative predictive value of 46.39% ± 0.01 and 60.45% (53.65% to 66.86% CI) and accuracy of 46.11% ± 0.01 and 57.78% (52.49% to 62.94% CI). When evaluating agreement across multiple images from the same patient, intra-reader agreement was 77.2% for AutoML and 90.8%-92.5% for the 3 radiologists. The models' F1 scores for the development and validation cohort were 0.78 and 0.62, respectively.

Conclusions

Liver fibrosis assessment in children is challenging without biopsy. An ultrasound-based AI model showed high sensitivity compared to radiologists, albeit still without suitable diagnostic performance for clinical use.

Open Access Review Article Issue
Renal Contrast-enhanced Ultrasound: Clinical Applications and Emerging Research
Advanced Ultrasound in Diagnosis and Therapy 2022, 6(4): 129-146
Published: 01 December 2022
Abstract PDF (1.6 MB) Collect
Downloads:15

Contrast-enhanced ultrasound (CEUS) is an imaging modality that has achieved considerable relevance in various clinical settings including the assessment of renal disease. CEUS is performed by injecting microbubble-based ultrasound contrast agents (UCAs) that create signals to display the microvasculature, allowing quantitative and qualitative assessment of parenchymal perfusion and real-time visualization of the renal anatomy. In recent years, CEUS has been widely accepted and applied for the assessment of kidney perfusion and the characterization of indeterminate renal masses, primarily due to its diagnostic efficacy, availability, low cost, reproducibility, and absence of nephrotoxicity. CEUS provides a higher spatial and temporal resolution than other cross-sectional imaging, resulting in high sensitivity and specificity for its applications in a variety of renal conditions including cancer monitoring following ablation, detection of transplant complications, hypoperfusion, acute traumatic injury, renal artery stenosis, parenchymal infection, and kidney intervention guidance. Additionally, the continuous investigation and development of new technologies surrounding this imaging technique have shown encouraging preliminary results for the use of CEUS in the evaluation of molecular expression in several disease processes, the dynamic analysis of blood flow kinetics, and the implementation of super-resolution imaging systems. The purpose of this article is to provide an overview of the current and potential clinical applications of renal CEUS.

Open Access Editorial Commentary Issue
Novel Development of Ultrasound Tomography for Musculoskeletal Imaging
Advanced Ultrasound in Diagnosis and Therapy 2024, 8(1): 39
Published: 30 March 2024
Abstract PDF (993.8 KB) Collect
Downloads:149
Open Access Review Article Issue
Emerging Applications of Contrast-enhanced Ultrasound in Trauma
Advanced Ultrasound in Diagnosis and Therapy 2022, 6(2): 39-47
Published: 30 June 2022
Abstract PDF (3.1 MB) Collect
Downloads:5

The use of contrast-enhanced ultrasound (CEUS) has expanded over the past decade to include a variety of diagnostic and therapeutic applications. These include urgent clinical situations that require timely diagnosis and subsequent treatment. With the introduction of microbubble ultrasound contrast agents (UCAs), CEUS provides increased sensitivity and specificity over conventional ultrasound. Within the trauma setting, CEUS benefits include point of care imaging and an ability to monitor perfusion in real-time. Additionally, UCAs are non-nephrotoxic, and can be used when contrast enhanced CT is contraindicated. In this review, we discuss recent advancements of CEUS within trauma settings.

Open Access Case Report Issue
Contrast-enhanced Ultrasound Assessment of Treatment Response in a Patient with Multifocal Hepatocellular Carcinoma Treated with Transarterial Chemo and Radioembolization
Advanced Ultrasound in Diagnosis and Therapy 2021, 5(3): 254-257
Published: 30 September 2021
Abstract PDF (10.5 MB) Collect
Downloads:2

Minimally invasive locoregional therapies have become important treatment options for patients with intermediate or late-stage hepatocellular carcinoma (HCC) who are ineligible for surgical resection or liver transplantation. Imaging modalities are essential for procedural guidance and for assessing treatment response thereafter. We report a unique finding of a patient with multifocal HCC treated with transarterial radioembolization (TARE) with yttrium-90 (Y90) and transarterial chemoembolization (TACE). We compared contrast-enhanced ultrasound (CEUS) with contrast-enhanced magnetic resonance imaging (CE-MRI) in the evaluation of treatment response to demonstrate advantages of CEUS imaging technique and its early detection of viable tumor.

Open Access Review Article Issue
Recent Advances in Microbubble-Augmented Cancer Therapy
Advanced Ultrasound in Diagnosis and Therapy 2020, 4(3): 155-168
Published: 30 August 2020
Abstract PDF (1.6 MB) Collect
Downloads:28

Contrast-enhanced ultrasound (CEUS) applications in cancer management have expanded over the past two decades. Through detection of vascularization and perfusion changes, CEUS provides a potentially reliable means of early prediction of response to different cancer therapies including systemic chemotherapy and locoregional therapies. Ultrasound-induced cavitation of contrast agents has a range of effects on the surrounding microenvironment. These effects can be manipulated to sensitize the tumors to radio- and chemotherapy, as well as achieve targeted delivery through drug-loaded contrast agents. Newer forms of drug carriers are being developed with improved drug-carrying capacity and tissue penetration. This review aims at providing a synopsis of the latest developments in CEUS’ use in oncologic therapy. While the majority of work described in this review is still in the pre-clinical phases, results have been encouraging and show potential translational benefit for cancer patients in the near future.

Open Access Original Research Issue
Automated Machine Learning in the Sonographic Diagnosis of Non-alcoholic Fatty Liver Disease
Advanced Ultrasound in Diagnosis and Therapy 2020, 4(3): 176-182
Published: 30 August 2020
Abstract PDF (1.2 MB) Collect
Downloads:8
Objective

This study evaluated the performance of automated machine-learning to diagnose non-alcoholic fatty liver disease (NAFLD) by ultrasound and compared these findings to radiologist performance.

Methods

96 patients with histologic (33) or proton density fat fraction MRI (63) diagnosis of NAFLD and 100 patients without evidence of NAFLD were retrospectively identified. The “Fatty Liver” label included 96 patients with 405 images and the “Not Fatty Liver” label included 100 patients with 500 images. These 905 images made up a “Comprehensive Image” group. A “Radiology Selected Image” group was then created by selecting only images considered diagnostic by a blinded radiologist, resulting in 649 images. Cloud AutoML Visionbeta (Google LLC, Mountain View, CA) was used for machine learning. The models were evaluated against three blinded radiologists.

Results

The “Comprehensive Image” group model demonstrated a sensitivity of 88.6% (73.3–96.8%) and a specificity of 95.3% (84.2–99.4%). Radiologist performance on this image group included a sensitivity of 81.0% (74.3–87.6%) and specificity of 86.0% (72.6–99.5%). The model’s overall accuracy was 92.3% (84.0–97.1%), compared with mean individual performance (83.8%, 78.4–89.1%). The “Radiology Selected Image” group model demonstrated a sensitivity of 88.6% (73.3 – 96.8%) and specificity of 87.9% (71.8–96.6%). Mean radiologist sensitivity was 92.4% (86.9–97.9%) and specificity was 91.9% (83.4–100%). The model’s overall accuracy was 88.2% (78.1–94.8%) which was comparable to the individual radiologist performance (92.2%, 90.1–94.2%) and consensus performance (95.6%, 87.6–99.1%).

Conclusions

An automated machine-learning algorithm may accurately detect NAFLD on ultrasound.

Open Access Review Article Issue
Applications in Molecular Ultrasound Imaging: Present and Future
Advanced Ultrasound in Diagnosis and Therapy 2019, 3(3): 62-75
Published: 30 September 2019
Abstract PDF (1.9 MB) Collect
Downloads:11

Molecular ultrasound imaging or targeted contrast-enhanced ultrasound (CEUS) is a relatively new technique that has varied applications to augment both diagnostics and therapeutics. Ultrasound contrast agents are conjugated to ligands that bind with specific biomarkers in the areas of interest which can then be quantified using ultrasound technology. This technique has numerous clinical applications including studying pathophysiology of disease, improving diagnostic sensitivity and specificity, and improving localized drug delivery. This technology, most notably, has proven useful in numerous oncologic and cardiovascular applications. Given ultrasound’s advantages over other radiographic studies including its low cost, lack of ionizing radiation, portability, ability to provide real-time imaging, and non-invasiveness, recent investigations have expanded the utility of molecular ultrasound. In this review, we briefly review targeted ultrasound contrast agents and explore the current applications of molecular ultrasound as well as future applications based on the currently published literature.

Open Access Review Article Issue
Artificial Intelligence in Ultrasound Imaging: Current Research and Applications
Advanced Ultrasound in Diagnosis and Therapy 2019, 3(3): 53-61
Published: 30 September 2019
Abstract PDF (1.6 MB) Collect
Downloads:12

Artificial intelligence (AI) is an area of computer science that emphasizes the creation of intelligent software or system based on big data information, machine learning and deep learning technologies. The rapid development of science and technology as well as internet communication has enabled AI and big data to gradually apply to many fields of health care. The modern imaging medicine is one of the first areas where AI can play an important role and applications. As cross-sectional imaging, ultrasound (US) is well suitable for AI technology to standardize imaging protocols and improve diagnostic accuracy. This article reviews current AI technology and related clinical applications in the fields of thyroid, breast and liver US.

Total 9