Zobrazeno 1 - 10
of 1 630
pro vyhledávání: '"Rubin, Daniel A."'
Autor:
Saab, Khaled, Tang, Siyi, Taha, Mohamed, Lee-Messer, Christopher, Ré, Christopher, Rubin, Daniel
A major barrier to deploying healthcare AI models is their trustworthiness. One form of trustworthiness is a model's robustness across different subgroups: while existing models may exhibit expert-level performance on aggregate metrics, they often re
Externí odkaz:
http://arxiv.org/abs/2306.08728
Autor:
Timcheck, Jonathan, Shrestha, Sumit Bam, Rubin, Daniel Ben Dayan, Kupryjanow, Adam, Orchard, Garrick, Pindor, Lukasz, Shea, Timothy, Davies, Mike
A critical enabler for progress in neuromorphic computing research is the ability to transparently evaluate different neuromorphic solutions on important tasks and to compare them to state-of-the-art conventional solutions. The Intel Neuromorphic Dee
Externí odkaz:
http://arxiv.org/abs/2303.09503
Autor:
van der Sluijs, Rogier, Bhaskhar, Nandita, Rubin, Daniel, Langlotz, Curtis, Chaudhari, Akshay
Publikováno v:
Proceedings of Machine Learning Research, MIDL 2023
Image augmentations are quintessential for effective visual representation learning across self-supervised learning techniques. While augmentation strategies for natural imaging have been studied extensively, medical images are vastly different from
Externí odkaz:
http://arxiv.org/abs/2301.12636
Autor:
Tang, Siyi, Dunnmon, Jared A., Qu, Liangqiong, Saab, Khaled K., Baykaner, Tina, Lee-Messer, Christopher, Rubin, Daniel L.
Multivariate biosignals are prevalent in many medical domains, such as electroencephalography, polysomnography, and electrocardiography. Modeling spatiotemporal dependencies in multivariate biosignals is challenging due to (1) long-range temporal dep
Externí odkaz:
http://arxiv.org/abs/2211.11176
Autor:
Mirzazadeh, Ali, Dubost, Florian, Pike, Maxwell, Maniar, Krish, Zuo, Max, Lee-Messer, Christopher, Rubin, Daniel
Attention--or attribution--maps methods are methods designed to highlight regions of the model's input that were discriminative for its predictions. However, different attention maps methods can highlight different regions of the input, with sometime
Externí odkaz:
http://arxiv.org/abs/2210.09705
Autor:
Alam, Minhaj Nur, Yamashita, Rikiya, Ramesh, Vignav, Prabhune, Tejas, Lim, Jennifer I., Chan, R. V. P., Hallak, Joelle, Leng, Theodore, Rubin, Daniel
Self supervised contrastive learning based pretraining allows development of robust and generalized deep learning models with small, labeled datasets, reducing the burden of label generation. This paper aims to evaluate the effect of CL based pretrai
Externí odkaz:
http://arxiv.org/abs/2208.11563
Publikováno v:
IEEE Transactions on Artificial Intelligence, 2023
Continuous monitoring of trained ML models to determine when their predictions should and should not be trusted is essential for their safe deployment. Such a framework ought to be high-performing, explainable, post-hoc and actionable. We propose TRU
Externí odkaz:
http://arxiv.org/abs/2207.11290
Domain generalization in medical image classification is an important problem for trustworthy machine learning to be deployed in healthcare. We find that existing approaches for domain generalization which utilize ground-truth abnormality segmentatio
Externí odkaz:
http://arxiv.org/abs/2206.08794
Autor:
Yan, Rui, Qu, Liangqiong, Wei, Qingyue, Huang, Shih-Cheng, Shen, Liyue, Rubin, Daniel, Xing, Lei, Zhou, Yuyin
The collection and curation of large-scale medical datasets from multiple institutions is essential for training accurate deep learning models, but privacy concerns often hinder data sharing. Federated learning (FL) is a promising solution that enabl
Externí odkaz:
http://arxiv.org/abs/2205.08576
Autor:
Yan-Ran, Wang, Qu, Liangqiong, Sheybani, Natasha Diba, Luo, Xiaolong, Wang, Jiangshan, Hawk, Kristina Elizabeth, Theruvath, Ashok Joseph, Gatidis, Sergios, Xiao, Xuerong, Pribnow, Allison, Rubin, Daniel, Daldrup-Link, Heike E.
Despite its tremendous value for the diagnosis, treatment monitoring and surveillance of children with cancer, whole body staging with positron emission tomography (PET) is time consuming and associated with considerable radiation exposure. 100x (1%
Externí odkaz:
http://arxiv.org/abs/2205.04044