BenchMD: A Benchmark for Unified Learning on Medical Images and Sensors
Authors:
Kathryn Wantlin,
Chenwei Wu,
Shih-Cheng Huang,
Oishi Banerjee,
Farah Dadabhoy,
Veeral Vipin Mehta,
Ryan Wonhee Han,
Fang Cao,
Raja R. Narayan,
Errol Colak,
Adewole Adamson,
Laura Heacock,
Geoffrey H. Tison,
Alex Tamkin,
Pranav Rajpurkar
Abstract:
Medical data poses a daunting challenge for AI algorithms: it exists in many different modalities, experiences frequent distribution shifts, and suffers from a scarcity of examples and labels. Recent advances, including transformers and self-supervised learning, promise a more universal approach that can be applied flexibly across these diverse conditions. To measure and drive progress in this dir…
▽ More
Medical data poses a daunting challenge for AI algorithms: it exists in many different modalities, experiences frequent distribution shifts, and suffers from a scarcity of examples and labels. Recent advances, including transformers and self-supervised learning, promise a more universal approach that can be applied flexibly across these diverse conditions. To measure and drive progress in this direction, we present BenchMD: a benchmark that tests how well unified, modality-agnostic methods, including architectures and training techniques (e.g. self-supervised learning, ImageNet pretraining),perform on a diverse array of clinically-relevant medical tasks. BenchMD combines 19 publicly available datasets for 7 medical modalities, including 1D sensor data, 2D images, and 3D volumetric scans. Our benchmark reflects real-world data constraints by evaluating methods across a range of dataset sizes, including challenging few-shot settings that incentivize the use of pretraining. Finally, we evaluate performance on out-of-distribution data collected at different hospitals than the training data, representing naturally-occurring distribution shifts that frequently degrade the performance of medical AI models. Our baseline results demonstrate that no unified learning technique achieves strong performance across all modalities, leaving ample room for improvement on the benchmark. Code is released at https://github.com/rajpurkarlab/BenchMD.
△ Less
Submitted 26 June, 2023; v1 submitted 17 April, 2023;
originally announced April 2023.
Interfacial Ferromagnetism in LaNiO3/CaMnO3 Superlattices
Authors:
A. J. Grutter,
H. Yang,
B. J. Kirby,
M. R. Fitzsimmons,
J. A. Aguiar,
N. D. Browning,
C. A. Jenkins,
E. Arenholz,
V. V. Mehta,
U. S. Alaan,
Y. Suzuki
Abstract:
We observe interfacial ferromagnetism in superlattices of the paramagnetic metal LaNiO3 and the antiferromagnetic insulator CaMnO3. LaNiO3 exhibits a thickness dependent metal-insulator transition and we find the emergence of ferromagnetism to be coincident with the conducting state of LaNiO3. That is, only superlattices in which the LaNiO3 layers are metallic exhibit ferromagnetism. Using several…
▽ More
We observe interfacial ferromagnetism in superlattices of the paramagnetic metal LaNiO3 and the antiferromagnetic insulator CaMnO3. LaNiO3 exhibits a thickness dependent metal-insulator transition and we find the emergence of ferromagnetism to be coincident with the conducting state of LaNiO3. That is, only superlattices in which the LaNiO3 layers are metallic exhibit ferromagnetism. Using several magnetic probes, we have determined that the ferromagnetism arises in a single unit cell of CaMnO3 at the interface. Together these results suggest that ferromagnetism can be attributed to a double exchange interaction among Mn ions mediated by the adjacent itinerant metal.
△ Less
Submitted 30 July, 2013;
originally announced July 2013.
