Yu-hui Chen
I joined Google at the end of 2015, starting with Smart Imagery team in Geo. Since Feb 2018, I am a member of Medical Brain and Health Research. My works mainly focus on extracting structural information from images/texts and building up infrastructure for powering the development of ML models. My area of interest include computer vision, natural language processing, signal processing, and machine learning.
I received my Ph.D. in Electrical and Computer Engineering from University of Michigan, working with Prof. Alfred Hero on multi-modality image fusion in biomedical and materials in 2015. The main idea of image fusion is to intelligently combine the information from multiple image modalities for better decision making, e.g. detecting anomaly. I received my B.S. in Electrical Engineering from National Taiwan University from 2005 to 2009, working with Prof. Lin-Shan Lee on speech recognition and information retrieval.
Personal:
My hometown is Taipei, Taiwan, where I was born and grew up. I came to United States in 2010 for my graduate degree and started to learn and enjoy snowboarding since then. I love music (classical, Rock, and R&B) and play a little bit piano and drums. Badminton is my favorite sport which I play regularly.
I received my Ph.D. in Electrical and Computer Engineering from University of Michigan, working with Prof. Alfred Hero on multi-modality image fusion in biomedical and materials in 2015. The main idea of image fusion is to intelligently combine the information from multiple image modalities for better decision making, e.g. detecting anomaly. I received my B.S. in Electrical Engineering from National Taiwan University from 2005 to 2009, working with Prof. Lin-Shan Lee on speech recognition and information retrieval.
Personal:
My hometown is Taipei, Taiwan, where I was born and grew up. I came to United States in 2010 for my graduate degree and started to learn and enjoy snowboarding since then. I love music (classical, Rock, and R&B) and play a little bit piano and drums. Badminton is my favorite sport which I play regularly.
Research Areas
Authored Publications
Sort By
The Medical Scribe: Corpus Development and Model Performance Analyses
Amanda Perry
Ashley Robson Domin
Chris Co
Gang Li
Hagen Soltau
Justin Stuart Paul
Lauren Keyes
Linh Tran
Mark David Knichel
Mingqiu Wang
Nan Du
Rayman Huang
Proc. Language Resources and Evaluation, 2020
Preview abstract
There has been a growing interest in creating tools to assist clinical note generation from the audio of provider-patient encounters. Motivated by this goal and with the help of providers and experienced medical scribes, we developed an annotation scheme to extract relevant clinical concepts. Using this annotation scheme, a corpus of about 6k clinical encounters was labeled, which was used to train a state-of-the-art tagging model. We report model performance and a detailed analyses of the results.
View details
Extracting Symptoms and their Status from Clinical Conversations
Nan Du
Linh Tran
Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics, Florence, Italy (2019), pp. 915-9125
Preview abstract
This paper describes novel models tailored for a new application, that of extracting the symptoms mentioned in clinical conversations along with their status. Lack of any publicly available corpus in this privacy-sensitive domain led us to develop our own corpus, consisting of about 3K conversations annotated by professional medical scribes. We propose two novel deep learning approaches to infer the symptom names and their status: (1) a new hierarchical span-attribute tagging (SAT) model, trained using curriculum learning, and (2) a variant of sequence-to-sequence model which decodes the symptoms and their status from a few speaker turns within a sliding window over the conversation. This task stems from a realistic application of assisting medical providers in capturing symptoms mentioned by patients from their clinical conversations. To reflect this application, we define multiple metrics. From inter-rater agreement, we find that the task is inherently difficult. We conduct comprehensive evaluations on several contrasting conditions and observe that the performance of the models range from an F-score of 0.5 to 0.8 depending on the condition. Our analysis not only reveals the inherent challenges of the task, but also provides useful directions to improve the models.
View details