There is great interest in developing radiological classifiers for diagnosis, staging, and predictive modeling in progressive diseases such as Parkinson's disease (PD), a neurodegenerative disease ...that is difficult to detect in its early stages. Here we leverage severity-based meta-data on the stages of disease to define a curriculum for training a deep convolutional neural network (CNN). Typically, deep learning networks are trained by randomly selecting samples in each mini-batch. By contrast, curriculum learning is a training strategy that aims to boost classifier performance by starting with examples that are easier to classify. Here we define a curriculum to progressively increase the difficulty of the training data corresponding to the Hoehn and Yahr (H&Y) staging system for PD (total N=1,012; 653 PD patients, 359 controls; age range: 20.0-84.9 years). Even with our multi-task setting using pre-trained CNNs and transfer learning, PD classification based on T1-weighted (T1-w) MRI was challenging (ROC AUC: 0.59-0.65), but curriculum training boosted performance (by 3.9%) compared to our baseline model. Future work with multimodal imaging may further boost performance.
Exoplanet discoveries have revealed a dramatic diversity of planet sizes across a vast array of orbital architectures. Sub-Neptunes are of particular interest; due to their absence in our own solar ...system, we rely on demographics of exoplanets to better understand their bulk composition and formation scenarios. Here, we present the discovery and characterization of TOI-1437 b, a sub-Neptune with a 18.84 day orbit around a near-Solar analog (Mstar = 1.10 +/- 0.10 Msun, Rstar = 1.17 +/- 0.12 Rsun). The planet was detected using photometric data from the Transiting Exoplanet Survey Satellite (TESS) mission and radial velocity follow-up observations were carried out as a part of the TESS-Keck Survey (TKS) using both the HIRES instrument at Keck Observatory and the Levy Spectrograph on the Automated Planet Finder (APF) telescope. A combined analysis of these data reveal a planet radius of Rp = 2.24 +/- 0.23 Rearth and a mass measurement of Mp = 9.6 +/- 3.9 Mearth). TOI-1437 b is one of few (~50) known transiting sub-Neptunes orbiting a solar-mass star that has a radial velocity mass measurement. As the formation pathway of these worlds remains an unanswered question, the precise mass characterization of TOI-1437 b may provide further insight into this class of planet.
Colonialism occurs when one group exerts control over people and territories outside its boundaries; this domination is based on both the colonizer’s perceptions of the relative value of populations ...and the intentional labor control, oppression, and economic marginalization of the colonized (Gasco 2005; Schortman and Urban 1998; Silliman 2005; Wernke and Whitmore 2009). The agency of the colonized varies depending on political organization, individual sociopolitical power, and circumstance (Alexander and Kepecs 2005; Ruiz Medrano 2010; Silliman 2005).
The Viceroyalty period of New Spain spanned from just after the Cortés expedition in 1535 to the Mexican revolt and subsequent independence from
Postmortem MRI allows brain anatomy to be examined at high resolution and to link pathology measures with morphometric measurements. However, automated segmentation methods for brain mapping in ...postmortem MRI are not well developed, primarily due to limited availability of labeled datasets, and heterogeneity in scanner hardware and acquisition protocols. In this work, we present a high resolution of 135 postmortem human brain tissue specimens imaged at 0.3 mm\(^{3}\) isotropic using a T2w sequence on a 7T whole-body MRI scanner. We developed a deep learning pipeline to segment the cortical mantle by benchmarking the performance of nine deep neural architectures, followed by post-hoc topological correction. We then segment four subcortical structures (caudate, putamen, globus pallidus, and thalamus), white matter hyperintensities, and the normal appearing white matter. We show generalizing capabilities across whole brain hemispheres in different specimens, and also on unseen images acquired at 0.28 mm^3 and 0.16 mm^3 isotropic T2*w FLASH sequence at 7T. We then compute localized cortical thickness and volumetric measurements across key regions, and link them with semi-quantitative neuropathological ratings. Our code, Jupyter notebooks, and the containerized executables are publicly available at: https://pulkit-khandelwal.github.io/exvivo-brain-upenn
Spinal cord ischemia (SCI) is an uncommon but serious complication of thoracic endovascular aortic repair (TEVAR). SCI after TEVAR is thought to result from decreased segmental blood supply to an ...important network of collateral blood flow in the spinal cord. Little is known about the prevalence and optimal treatment of SCI that occurs beyond the periprocedural period. We report a case of delayed SCI in a 67-year-old patient who underwent TEVAR. The patient presented almost two years after TEVAR with acute paraplegia preceded by pre-syncope. The delayed SCI was likely triggered by pre-syncope, a thrombosed endoleak shown on imaging, and the patient's vascular risk factors. Treatments included cerebrospinal fluid (CSF) drainage, mean arterial pressure (MAP) augmentation, and a naloxone infusion, which resulted in moderate recovery in lower extremity motor function. This case highlights the tenuous nature of spinal cord perfusion after TEVAR and that prompt recognition and early treatment of SCI are critical in preventing the progression from ischemia to infarction.
The approaches by which the machine learning and clinical research communities utilize real world data (RWD), including data captured in the electronic health record (EHR), vary dramatically. While ...clinical researchers cautiously use RWD for clinical investigations, ML for healthcare teams consume public datasets with minimal scrutiny to develop new algorithms. This study bridges this gap by developing and validating ML-DQA, a data quality assurance framework grounded in RWD best practices. The ML-DQA framework is applied to five ML projects across two geographies, different medical conditions, and different cohorts. A total of 2,999 quality checks and 24 quality reports were generated on RWD gathered on 247,536 patients across the five projects. Five generalizable practices emerge: all projects used a similar method to group redundant data element representations; all projects used automated utilities to build diagnosis and medication data elements; all projects used a common library of rules-based transformations; all projects used a unified approach to assign data quality checks to data elements; and all projects used a similar approach to clinical adjudication. An average of 5.8 individuals, including clinicians, data scientists, and trainees, were involved in implementing ML-DQA for each project and an average of 23.4 data elements per project were either transformed or removed in response to ML-DQA. This study demonstrates the importance role of ML-DQA in healthcare projects and provides teams a framework to conduct these essential activities.