Identifying Medical Diagnoses and Treatable Diseases by Image-Based Deep Learning.

The implementation of clinical-decision support algorithms for medical imaging faces challenges with reliability and interpretability. Here, we establish a diagnostic tool based on a deep-learning framework for the screening of patients with common treatable blinding retinal diseases. Our framework utilizes transfer learning, which trains a neural network with a fraction of the data of conventional approaches. Applying this approach to a dataset of optical coherence tomography images, we demonstrate performance comparable to that of human experts in classifying age-related macular degeneration and diabetic macular edema. We also provide a more transparent and interpretable diagnosis by highlighting the regions recognized by the neural network. We further demonstrate the general applicability of our AI system for diagnosis of pediatric pneumonia using chest X-ray images. This tool may ultimately aid in expediting the diagnosis and referral of these treatable conditions, thereby facilitating earlier treatment, resulting in improved clinical outcomes. VIDEO ABSTRACT.

Pharmacologically Improved Contractility Protects Against Aortic Dissection in Mice With Disrupted Transforming Growth Factor-ß Signaling Despite Compromised Extracellular Matrix Properties.

OBJECTIVE: Transforming growth factor-beta is a pleiotropic cytokine having diverse roles in vascular morphogenesis, homeostasis, and pathogenesis. Altered activity of and signaling through transforming growth factor-beta has been implicated in thoracic aortic aneurysms and dissections, conditions characterized by a reduced structural integrity of the wall that associates with altered biomechanics and mechanobiology. We quantify and contrast the passive and active biaxial biomechanical properties of the ascending and proximal descending thoracic aorta in a mouse model of altered transforming growth factor-beta signaling, with and without treatment with rapamycin. APPROACH AND RESULTS: Postnatal disruption of the gene (Tgfbr2) that codes the type II transforming growth factor-beta receptor compromises vessel-level contractility and elasticity. Daily treatment with rapamycin, a mechanistic target of rapamycin inhibitor that protects against aortic dissection in these mice, largely preserves or restores the contractile function while the passive properties remain compromised. Importantly, this increased smooth muscle contractility protects an otherwise vulnerable aortic wall from pressure-induced intramural delaminations in vitro. CONCLUSIONS: Notwithstanding the protection afforded by rapamycin in vivo and in vitro, the residual mechanical dysfunctionality suggests a need for caution if rapamycin is to be considered as a potential therapeutic. There is a need for in vivo evaluations in cases of increased hemodynamic loading, including hypertension or extreme exercise, which could unduly stress a structurally vulnerable aortic wall. Given these promising early results, however, such studies are clearly warranted.