Point-of-care AI-assisted stepwise ultrasound pneumothorax diagnosis

Objective. Ultrasound is extensively utilized as a convenient and cost-effective method in emergency situations. Unfortunately, the limited availability of skilled clinicians in emergency hinders the wider adoption of point-of-care ultrasound. To overcome this challenge, this paper aims to aid less experienced healthcare providers in emergency lung ultrasound scans.Approach. To assist healthcare providers, it is important to have a comprehensive model that can automatically guide the entire process of lung ultrasound based on the clinician's workflow. In this paper, we propose a framework for diagnosing pneumothorax using artificial intelligence (AI) assistance. Specifically, the proposed framework for lung ultrasound scan follows the steps taken by skilled physicians. It begins with finding the appropriate transducer position on the chest to locate the pleural line accurately in B-mode. The next step involves acquiring temporal M-mode data to determine the presence of lung sliding, a crucial indicator for pneumothorax. To mimic the sequential process of clinicians, two DL models were developed. The first model focuses on quality assurance (QA) and regression of the pleural line region-of-interest, while the second model classifies lung sliding. To achieve the inference on a mobile device, a size of EfficientNet-Lite0 model was further reduced to have fewer than 3 million parameters.Main results. The results showed that both the QA and lung sliding classification models achieved over 95% in area under the receiver operating characteristic (AUC), while the ROI performance reached 89% in the dice similarity coefficient. The entire stepwise pipeline was simulated using retrospective data, yielding an AUC of 89%.Significance. The step-wise AI framework for the pneumothorax diagnosis with QA offers an intelligible guide for each clinical workflow, which achieved significantly high precision and real-time inferences.

Deoxycholic Acid for Dercum Disease: Repurposing a Cosmetic Agent to Treat a Rare Disease

Dercum disease is a rare condition characterized by multiple painful fatty tumors distributed throughout the body. There currently are no US Food and Drug Administration-approved treatments for Dercum disease, and the treatments tried have shown little to no efficacy, leaving many patients with a profoundly negative impact on quality of life. We present a case series of 3 patients who were diagnosed with Dercum disease and were treated with deoxycholic acid (DCA), a therapy approved for adipolysis of submental fat. The patients experienced a reduction in tumor size with radiographic evidence as well as a notable reduction in symptoms.

Optimizing Lung Ultrasound: The Effect of Depth, Gain and Focal Position on Sonographic B-Lines

Ultrasonographic B-lines are artifacts present in alveolar-interstitial syndromes. We prospectively investigated optimal depth, gain, focal position and transducer type for B-line visualization and image quality. B-Lines were assessed at a single rib interspace with curvilinear and linear transducers. Video clips were acquired by changing parameters: depth (6, 12, 18 and 24 cm for curvilinear transducer, 4 and 8 cm for linear transducer), gain (10%, 50% and 90%) and focal position (at the pleural line or half the scanning depth). Clips were scored for B-lines and image quality. Five hundred sixteen clips were obtained and analyzed. The curvilinear transducer improved B-line visualization (63% vs. 37%, p < 0.0001), with higher image quality (3.52 ± 0.71 vs. 3.31 ± 0.86, p = 0.0047) compared with the linear transducer. B-Lines were better visualized at higher gains (curvilinear: gain of 50% vs. 10%, odds ratio = 7.04, 95% confidence interval: 4.03-12.3; gain of 90% vs. 10%, odds ratio = 9.48, 95% confidence interval: 5.28-17.0) and with the focal point at the pleural line (odds ratio = 1.64, 95% confidence interval: 1.02-2.63). Image quality was highest at 50% gain (p = 0.02) but decreased at 90% gain (p < 0.0001) and with the focal point at the pleural line (p < 0.0001). Image quality was highest at depths of 12-18 cm. B-Lines are best visualized using a curvilinear transducer with at least 50% gain and focal position at the pleural line. Gain less than 90% and image depth between 12 and 18 cm improve image quality.

Detecting Ventilator-Induced Diaphragmatic Dysfunction Using Point-of-Care Ultrasound in Children With Long-term Mechanical Ventilation

Long-term mechanical ventilation (MV) is defined as the use of MV for more than 6 hours per day for at least 3 weeks. Children requiring long-term MV include those with neuromuscular disease, central dysregulation, or lung dysfunction. Such children with medical complexity may be at risk for ventilator-induced diaphragmatic dysfunction. Ventilator-induced diaphragmatic dysfunction has been described in adult patients requiring acute MV with ultrasound (US). At this time, diaphragmatic US has not been evaluated in the pediatric post-acute care setting or incorporated into weaning strategies. We present 24 cases of children requiring long-term MV who underwent diaphragmatic US examinations to evaluate for ventilator-induced diaphragmatic dysfunction.