Application of deep learning algorithms in classification and localization of implant cutout for the postoperative hip.

OBJECTIVE: This study aims to explore the feasibility of employing convolutional neural networks for detecting and localizing implant cutouts on anteroposterior pelvic radiographs. MATERIALS AND METHODS: The research involves the development of two Deep Learning models. Initially, a model was created for image-level classification of implant cutouts using 40191 pelvic radiographs obtained from a single institution. The radiographs were partitioned into training, validation, and hold-out test datasets in a 6/2/2 ratio. Performance metrics including the area under the receiver operator characteristics curve (AUROC), sensitivity, and specificity were calculated using the test dataset. Additionally, a second object detection model was trained to localize implant cutouts within the same dataset. Bounding box visualizations were generated on images predicted as cutout-positive by the classification model in the test dataset, serving as an adjunct for assessing algorithm validity. RESULTS: The classification model had an accuracy of 99.7%, sensitivity of 84.6%, specificity of 99.8%, AUROC of 0.998 (95% CI: 0.996, 0.999) and AUPRC of 0.774 (95% CI: 0.646, 0.880). From the pelvic radiographs predicted as cutout-positive, the object detection model could achieve 95.5% localization accuracy on true positive images, but falsely generated 14 results from the 15 false-positive predictions. CONCLUSION: The classification model showed fair accuracy for detection of implant cutouts, while the object detection model effectively localized cutout. This serves as proof of concept of using a deep learning-based approach for classification and localization of implant cutouts from pelvic radiographs.

Déjà Vu or Jamais Vu? How the Severe Acute Respiratory Syndrome Experience Influenced a Singapore Radiology Department's Response to the Coronavirus Disease (COVID-19) Epidemic.

OBJECTIVE. This article shares the ground operational perspective of how a tertiary hospital radiology department in Singapore is responding to the coronavirus disease (COVID-19) epidemic. This same department was also deeply impacted by the severe acute respiratory syndrome (SARS) outbreak in 2003. CONCLUSION. Though similar to SARS, the COVID-19 outbreak has several differences. We share how lessons from 2003 are applied and modified in our ongoing operational response to this evolving novel pathogen.

Deep Learning Model and its Application for the Diagnosis of Exudative Pharyngitis.

OBJECTIVES: Telemedicine is firmly established in the healthcare landscape of many countries. Acute respiratory infections are the most common reason for telemedicine consultations. A throat examination is important for diagnosing bacterial pharyngitis, but this is challenging for doctors during a telemedicine consultation. A solution could be for patients to upload images of their throat to a web application. This study aimed to develop a deep learning model for the automated diagnosis of exudative pharyngitis. Thereafter, the model will be deployed online. METHODS: We used 343 throat images (139 with exudative pharyngitis and 204 without pharyngitis) in the study. ImageDataGenerator was used to augment the training data. The convolutional neural network models of MobileNetV3, ResNet50, and EfficientNetB0 were implemented to train the dataset, with hyperparameter tuning. RESULTS: All three models were trained successfully; with successive epochs, the loss and training loss decreased, and accuracy and training accuracy increased. The EfficientNetB0 model achieved the highest accuracy (95.5%), compared to MobileNetV3 (82.1%) and ResNet50 (88.1%). The EfficientNetB0 model also achieved high precision (1.00), recall (0.89) and F1-score (0.94). CONCLUSIONS: We trained a deep learning model based on EfficientNetB0 that can diagnose exudative pharyngitis. Our model was able to achieve the highest accuracy, at 95.5%, out of all previous studies that used machine learning for the diagnosis of exudative pharyngitis. We have deployed the model on a web application that can be used to augment the doctor's diagnosis of exudative pharyngitis.

Hyperbaric Oxygen Therapy Achieved Fistula Healing in a Young Patient With Severe Refractory Perianal Crohn's Disease.

The presence of perianal fistulae constitutes a more severe phenotype of Crohn's disease (CD) that often requires intensive medical therapy, wound care, and surgical intervention. Despite therapeutic advances in inflammatory bowel disease, the treatment of perianal fistulae remains challenging. Hyperbaric oxygen therapy (HBOT) has been proposed as an adjunctive treatment modality for induction of fistula healing. We illustrate a case in which HBOT achieved fistula healing in a young patient with severe refractory perianal Crohn's disease (pCD). We also review the current literature and discuss the role of HBOT in the treatment armamentarium of pCD.

Singapore tuberculosis (TB) clinical management guidelines 2024: A modified Delphi adaptation of international guidelines for drug-susceptible TB infection and pulmonary disease.

Introduction: Tuberculosis (TB) remains endemic in Singapore. Singapore's clinical practice guidelines for the management of tuberculosis were first published in 2016. Since then, there have been major new advances in the clinical management of TB, ranging from diagnostics to new drugs and treatment regimens. The National TB Programme convened a multidisciplinary panel to update guidelines for the clinical management of drug-susceptible TB infection and disease in Singapore, contextualising current evidence for local practice. Method: Following the ADAPTE framework, the panel systematically reviewed, scored and synthesised English-language national and international TB clinical guidelines published from 2016, adapting recommendations for a prioritised list of clinical decisions. For questions related to more recent advances, an additional primary literature review was conducted via a targeted search approach. A 2-round modified Delphi process was implemented to achieve consensus for each recommendation, with a final round of edits after consultation with external stakeholders. Results: Recommendations for 25 clinical questions spanning screening, diagnosis, selection of drug regimen, monitoring and follow-up of TB infection and disease were formulated. The availability of results from recent clinical trials led to the inclusion of shorter treatment regimens for TB infection and disease, as well as consensus positions on the role of newer technologies, such as computer-aided detection-artificial intelligence products for radiological screening of TB disease, next-generation sequencing for drug-susceptibility testing, and video observation of treatment. Conclusion: The panel updated recommendations on the management of drug-susceptible TB infection and disease in Singapore.

Erratum: Future Online Radiology Education: The Importance of Curriculum.

This corrects the article on p. 173 in vol. 24, PMID: 36788773.

Application of a deep learning algorithm in the detection of hip fractures.

This paper describes the development of a deep learning model for prediction of hip fractures on pelvic radiographs (X-rays). Developed using over 40,000 pelvic radiographs from a single institution, the model demonstrated high sensitivity and specificity when applied to a test set of emergency department radiographs. This study approximates the real-world application of a deep learning fracture detection model by including radiographs with sub-optimal image quality, other non-hip fractures, and metallic implants, which were excluded from prior published work. The study also explores the effect of ethnicity on model performance, as well as the accuracy of visualization algorithm for fracture localization.

A translational perspective towards clinical AI fairness.

Artificial intelligence (AI) has demonstrated the ability to extract insights from data, but the fairness of such data-driven insights remains a concern in high-stakes fields. Despite extensive developments, issues of AI fairness in clinical contexts have not been adequately addressed. A fair model is normally expected to perform equally across subgroups defined by sensitive variables (e.g., age, gender/sex, race/ethnicity, socio-economic status, etc.). Various fairness measurements have been developed to detect differences between subgroups as evidence of bias, and bias mitigation methods are designed to reduce the differences detected. This perspective of fairness, however, is misaligned with some key considerations in clinical contexts. The set of sensitive variables used in healthcare applications must be carefully examined for relevance and justified by clear clinical motivations. In addition, clinical AI fairness should closely investigate the ethical implications of fairness measurements (e.g., potential conflicts between group- and individual-level fairness) to select suitable and objective metrics. Generally defining AI fairness as "equality" is not necessarily reasonable in clinical settings, as differences may have clinical justifications and do not indicate biases. Instead, "equity" would be an appropriate objective of clinical AI fairness. Moreover, clinical feedback is essential to developing fair and well-performing AI models, and efforts should be made to actively involve clinicians in the process. The adaptation of AI fairness towards healthcare is not self-evident due to misalignments between technical developments and clinical considerations. Multidisciplinary collaboration between AI researchers, clinicians, and ethicists is necessary to bridge the gap and translate AI fairness into real-life benefits.

Multiparametric Magnetic Resonance Imaging and Magnetic Resonance Elastography to Evaluate the Early Effects of Bariatric Surgery on Nonalcoholic Fatty Liver Disease.

Background: Bariatric surgery is the most effective treatment for morbid obesity and reduces the severity of nonalcoholic fatty liver disease (NAFLD) in the long term. Less is known about the effects of bariatric surgery on liver fat, inflammation, and fibrosis during the early stages following bariatric surgery. Aims: This exploratory study utilises advanced imaging methods to investigate NAFLD and fibrosis changes during the early metabolic transitional period following bariatric surgery. Methods: Nine participants with morbid obesity underwent sleeve gastrectomy. Multiparametric MRI (mpMRI) and magnetic resonance elastography (MRE) were performed at baseline, during the immediate (1 month), and late (6 months) postsurgery period. Liver fat was measured using proton density fat fraction (PDFF), disease activity using iron-correct T1 (cT1), and liver stiffness using MRE. Repeated measured ANOVA was used to assess longitudinal changes and Dunnett's method for multiple comparisons. Results: All participants (Age 45.1 ± 9.0 years, BMI 39.7 ± 5.3 kg/m2) had elevated hepatic steatosis at baseline (PDFF >5%). In the immediate postsurgery period, PDFF decreased significantly from 14.1 ± 7.4% to 8.9 ± 4.4% (p = 0.016) and cT1 from 826.9 ± 80.6 ms to 768.4 ± 50.9 ms (p = 0.047). These improvements continued to the later postsurgery period. Bariatric surgery did not reduce liver stiffness measurements. Conclusion: Our findings support using MRI as a noninvasive tool to monitor NAFLD in patient with morbid obesity during the early stages following bariatric surgery.

Inferring cancer disease response from radiology reports using large language models with data augmentation and prompting.

OBJECTIVE: To assess large language models on their ability to accurately infer cancer disease response from free-text radiology reports. MATERIALS AND METHODS: We assembled 10 602 computed tomography reports from cancer patients seen at a single institution. All reports were classified into: no evidence of disease, partial response, stable disease, or progressive disease. We applied transformer models, a bidirectional long short-term memory model, a convolutional neural network model, and conventional machine learning methods to this task. Data augmentation using sentence permutation with consistency loss as well as prompt-based fine-tuning were used on the best-performing models. Models were validated on a hold-out test set and an external validation set based on Response Evaluation Criteria in Solid Tumors (RECIST) classifications. RESULTS: The best-performing model was the GatorTron transformer which achieved an accuracy of 0.8916 on the test set and 0.8919 on the RECIST validation set. Data augmentation further improved the accuracy to 0.8976. Prompt-based fine-tuning did not further improve accuracy but was able to reduce the number of training reports to 500 while still achieving good performance. DISCUSSION: These models could be used by researchers to derive progression-free survival in large datasets. It may also serve as a decision support tool by providing clinicians an automated second opinion of disease response. CONCLUSIONS: Large clinical language models demonstrate potential to infer cancer disease response from radiology reports at scale. Data augmentation techniques are useful to further improve performance. Prompt-based fine-tuning can significantly reduce the size of the training dataset.