Sociodemographic biases in a commercial AI model for intracranial hemorrhage detection.

PURPOSE: To evaluate whether a commercial AI tool for intracranial hemorrhage (ICH) detection on head CT exhibited sociodemographic biases. METHODS: Our retrospective study reviewed 9736 consecutive, adult non-contrast head CT scans performed between November 2021 and February 2022 in a single healthcare system. Each CT scan was evaluated by a commercial ICH AI tool and a board-certified neuroradiologist; ground truth was defined as final radiologist determination of ICH presence/absence. After evaluating the AI tool's aggregate diagnostic performance, sub-analyses based on sociodemographic groups (age, sex, race, ethnicity, insurance status, and Area of Deprivation Index [ADI] scores) assessed for biases. χ2 test or Fisher's exact tests evaluated for statistical significance with p ≤ 0.05. RESULTS: Our patient population was 50% female (mean age 60 ± 19 years). The AI tool had an aggregate accuracy of 93% [9060/9736], sensitivity of 85% [1140/1338], specificity of 94% [7920/ 8398], positive predictive value (PPV) of 71% [1140/1618] and negative predictive value (NPV) of 98% [7920/8118]. Sociodemographic biases were identified, including lower PPV for patients who were females (67.3% [62,441/656] vs. 72.7% [699/962], p = 0.02), Black (66.7% [454/681] vs. 73.2% [686/937], p = 0.005), non-Hispanic/non-Latino (69.7% [1038/1490] vs. 95.4% [417/437]), p = 0.009), and who had Medicaid/Medicare (69.9% [754/1078]) or Private (66.5% [228/343]) primary insurance (p = 0.003). Lower sensitivity was seen for patients in the third quartile of national (78.8% [241/306], p = 0.001) and state ADI scores (79.0% [22/287], p = 0.001). CONCLUSIONS: In our healthcare system, a commercial AI tool had lower performance for ICH detection than previously reported and demonstrated several sociodemographic biases.

Automated detection of incidental abdominal aortic aneurysms on computed tomography.

PURPOSE: To detect and assess abdominal aortic aneurysms (AAAs) on CT in a large asymptomatic adult patient population using fully-automated deep learning software. MATERIALS AND METHODS: The abdominal aorta was segmented using a fully-automated deep learning model trained on 66 manually-segmented abdominal CT scans from two datasets. The axial diameters of the segmented aorta were extracted to detect the presence of AAAs-maximum axial aortic diameter greater than 3 cm were labeled as AAA positive. The trained system was then externally-validated on CT colonography scans of 9172 asymptomatic outpatients (mean age, 57 years) referred for colorectal cancer screening. Using a previously-validated automated calcified atherosclerotic plaque detector, we correlated abdominal aortic Agatston and volume scores with the presence of AAA. RESULTS: The deep learning software detected AAA on the external validation dataset with a sensitivity, specificity, and AUC of 96%, (95% CI 89%, 100%), 96% (96%, 97%), and 99% (98%, 99%) respectively. The Agatston and volume scores of reported AAA-positive cases were statistically significantly greater than those of reported AAA-negative cases (p < 0.0001). Using plaque alone as a AAA detector, at a threshold Agatston score of 2871, the sensitivity and specificity were 84% (73%, 94%) and 87% (86%, 87%), respectively. CONCLUSION: Fully-automated detection and assessment of AAA on CT is feasible and accurate. There was a strong statistical association between the presence of AAA and the quantity of abdominal aortic calcified atherosclerotic plaque.

Children Are Not Small Adults: Addressing Limited Generalizability of an Adult Deep Learning CT Organ Segmentation Model to the Pediatric Population.

Deep learning (DL) tools developed on adult data sets may not generalize well to pediatric patients, posing potential safety risks. We evaluated the performance of TotalSegmentator, a state-of-the-art adult-trained CT organ segmentation model, on a subset of organs in a pediatric CT dataset and explored optimization strategies to improve pediatric segmentation performance. TotalSegmentator was retrospectively evaluated on abdominal CT scans from an external adult dataset (n = 300) and an external pediatric data set (n = 359). Generalizability was quantified by comparing Dice scores between adult and pediatric external data sets using Mann-Whitney U tests. Two DL optimization approaches were then evaluated: (1) 3D nnU-Net model trained on only pediatric data, and (2) an adult nnU-Net model fine-tuned on the pediatric cases. Our results show TotalSegmentator had significantly lower overall mean Dice scores on pediatric vs. adult CT scans (0.73 vs. 0.81, P < .001) demonstrating limited generalizability to pediatric CT scans. Stratified by organ, there was lower mean pediatric Dice score for four organs (P < .001, all): right and left adrenal glands (right adrenal, 0.41 [0.39-0.43] vs. 0.69 [0.66-0.71]; left adrenal, 0.35 [0.32-0.37] vs. 0.68 [0.65-0.71]); duodenum (0.47 [0.45-0.49] vs. 0.67 [0.64-0.69]); and pancreas (0.73 [0.72-0.74] vs. 0.79 [0.77-0.81]). Performance on pediatric CT scans improved by developing pediatric-specific models and fine-tuning an adult-trained model on pediatric images where both methods significantly improved segmentation accuracy over TotalSegmentator for all organs, especially for smaller anatomical structures (e.g., > 0.2 higher mean Dice for adrenal glands; P < .001).

Pre-operative Machine Learning for Heart Transplant Patients Bridged with Temporary Mechanical Circulatory Support.

Background: Existing prediction models for post-transplant mortality in patients bridged to heart transplantation with temporary mechanical circulatory support (tMCS) perform poorly. A more reliable model would allow clinicians to provide better pre-operative risk assessment and develop more targeted therapies for high-risk patients. Methods: We identified adult patients in the United Network for Organ Sharing database undergoing isolated heart transplantation between 01/2009 and 12/2017 who were supported with tMCS at the time of transplant. We constructed a machine learning model using extreme gradient boosting (XGBoost) with a 70:30 train:test split to predict 1-year post-operative mortality. All pre-transplant variables available in the UNOS database were included to train the model. Shapley Additive Explanations was used to identify and interpret the most important features for XGBoost predictions. Results: A total of 1584 patients were included, with a median age of 56 (interquartile range: 46-62) and 74% male. Actual 1-year mortality was 12.1%. Out of 498 available variables, 43 were selected for the final model. The area under the receiver operator characteristics curve (AUC) for the XGBoost model was 0.71 (95% CI: 0.62-0.78). The most important variables predictive of 1-year mortality included recipient functional status, age, pulmonary capillary wedge pressure (PCWP), cardiac output, ECMO usage, and serum creatinine. Conclusions: An interpretable machine learning model trained on a large clinical database demonstrated good performance in predicting 1-year mortality for patients bridged to heart transplantation with tMCS. Machine learning may be used to enhance clinician judgement in the care of markedly high-risk transplant recipients.

Perivascular fat attenuation for predicting adverse cardiac events in stable patients undergoing invasive coronary angiography.

BACKGROUND: Inflammation surrounding the coronary arteries can be non-invasively assessed using pericoronary adipose tissue attenuation (PCAT). While PCAT holds promise for further risk stratification of patients with low coronary artery disease (CAD) prevalence, its value in higher risk populations remains unknown. METHODS: CORE320 enrolled patients referred for invasive coronary angiography with known or suspected CAD. Coronary computed tomography angiography (CCTA) images were collected for 381 patients for whom clinical outcomes were assessed 5 years after enrollment. Using semi-automated image analysis software, PCAT was obtained and normalized for the right coronary (RCA), left anterior descending (LAD), and left circumflex arteries (LCx). The association between PCAT and major adverse cardiovascular events (MACE) during follow up was assessed using Cox regression models. RESULTS: Thirty-seven patients were excluded due to technical failure. For the remaining 344 patients, median age was 62 (interquartile range, 55-68) with 59% having ≥1 coronary artery stenosis of ≥50% by quantitative coronary angiography. Mean attenuation values for PCAT in RCA, LAD, and LCx were -74.9, -74.2, and -71.2, respectively. Hazard ratios and 95% confidence intervals (CI) for normalized PCAT in the RCA, LAD, and LCx for MACE were 0.96 (CI: 0.75-1.22, p â€‹= â€‹0.71), 1.31 (95% CI: 0.96-1.78, p â€‹= â€‹0.09), and 0.98 (95% CI: 0.78-1.22, p â€‹= â€‹0.84), respectively. For death, stroke, or myocardial infarction only, hazard ratios were 0.68 (0.44-1.07), 0.85 (0.56-1.29), and 0.57 (0.41-0.80), respectively. CONCLUSIONS: In patients referred for invasive coronary angiography with suspected CAD, PCAT did not predict MACE during long term follow up. Further studies are needed to understand the relationship of PCAT with CAD risk.

Direct Ink Writing of Poly(tetrafluoroethylene) (PTFE) with Tunable Mechanical Properties.

Poly(tetrafluoroethylene) (PTFE) is a unique polymer with highly desirable properties such as resistance to chemical degradation, biocompatibility, hydrophobicity, antistiction, and low friction coefficient. However, due to its high melt viscosity, it is not possible to three-dimensional (3D)-print PTFE structures using nozzle-based extrusion. Here, we report a new and versatile strategy for 3D-printing PTFE structures using direct ink writing (DIW). Our approach is based on a newly formulated PTFE nanoparticle ink and thermal treatment process. The ink was formulated by mixing an aqueous dispersion of surfactant-stabilized PTFE nanoparticles with a binding gum to optimize its shear-thinning properties required for DIW. We developed a multistage thermal treatment to fuse the PTFE nanoparticles, solidify the printed structures, and remove the additives. We have extensively characterized the rheological and mechanical properties and processing parameters of these structures using imaging, mechanical testing, and statistical design of experiments. Importantly, several of the mechanical and structural properties of the final-printed PTFE structures resemble that of compression-molded PTFE, and additionally, the mechanical properties are tunable. We anticipate that this versatile approach facilitates the production of 3D-printed PTFE components using DIW with significant potential applications in engineering and medicine.

Dual-Gel 4D Printing of Bioinspired Tubes.

The distribution of periodic patterns of materials with radial or bilateral symmetry is a universal natural design principle. Among the many biological forms, tubular shapes are a common motif in many organisms, and they are also important for bioimplants and soft robots. However, the simple design principle of strategic placement of 3D printed segments of swelling and nonswelling materials to achieve widely different functionalities is yet to be demonstrated. Here, we report the design, fabrication, and characterization of segmented 3D printed gel tubes composed of an active thermally responsive swelling gel (poly N-isopropylacrylamide) and a passive thermally nonresponsive gel (polyacrylamide). Using finite element simulations and experiments, we report a variety of shape changes including uniaxial elongation, radial expansion, bending, and gripping based on two gels. Actualization and characterization of thermally induced shape changes are of key importance to robotics and biomedical engineering. Our studies present rational approaches to engineer complex parameters with a high level of customization and tunability for additive manufacturing of dynamic gel structures.