Alexander T. M. Cheung

Alexander T. M. Cheung

Auto-generated profile, based on publicly available information

Alexander T. M. Cheung

NYU Grossman School of Medicine, New York, USA

Research Focus & Metrics

Academic Career Length

13Years (2014-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
7 June 2023 in Nature
Verified
Journal Article
Open access
PDF file
LJLavender Yao Jiang
Lavender Yao Jiang
XLXujin Chris Liu
Xujin Chris Liu
NNNima Pour Nejatian
Nima Pour Nejatian
MNMustafa Nasir-Moin
Mustafa Nasir-Moin
DWDuo Wang
Duo Wang
EOEric Karl Oermann
Eric Karl Oermann
Physicians make critical time-constrained decisions every day. Clinical predictive models can help physicians and administrators make decisions by forecasting clinical and operational events. Existing structured data-based clinical predictive models have limited use in everyday practice owing to complexity in data processing, as well as model development and deployment1–3. Here we show that unstructured clinical notes from the electronic health record can enable the training of clinical language models, which can be used as all-purpose clinical predictive engines with low-resistance development and deployment. Our approach leverages recent advances in natural language processing4,5to train a large language model for medical language (NYUTron) and subsequently fine-tune it across a wide range of clinical and operational predictive tasks. We evaluated our approach within our health system for five such tasks: 30-day all-cause readmission prediction, in-hospital mortality prediction, comorbidity index prediction, length of stay prediction, and insurance denial prediction. We show that NYUTron has an area under the curve (AUC) of 78.7–94.9%, with an improvement of 5.36–14.7% in the AUC compared with traditional models. We additionally demonstrate the benefits of pretraining with clinical text, the potential for increasing generalizability to different sites through fine-tuning and the full deployment of our system in a prospective, single-arm trial. These results show the potential for using clinical language models in medicine to read alongside physicians and provide guidance at the point of care.

1 May 2014 in Annals of Biomedical Engineering
Verified
Journal Article
Closed access
PDF file
MHMichael S. Hu
Michael S. Hu
ZMZeshaan Maan
Zeshaan Maan
JWJen-Chieh Wu
Jen-Chieh Wu
RRRobert C. Rennert
Robert C. Rennert
WHWan Xing Hong
Wan Xing Hong
HLH. Peter Lorenz
H. Peter Lorenz

2 January 2018 in Organogenesis
Verified
Journal Article
Closed access
MHMichael S. Hu
Michael S. Hu
MBMimi R. Borrelli
Mimi R. Borrelli
WHWan Xing Hong
Wan Xing Hong
SMSamir Malhotra
Samir Malhotra
ACAlexander T. M. Cheung
Alexander T. M. Cheung
Michael T. LongakerMichael T. Longaker
Michael T. Longaker
Fetal cutaneous wounds have the unique ability to completely regenerate wounded skin and heal without scarring. However, adult cutaneous wounds heal via a fibroproliferative response which results in the formation of a scar. Understanding the mechanism(s) of scarless wound healing leads to enormous clinical potential in facilitating an environment conducive to scarless healing in adult cutaneous wounds. This article reviews the embryonic development of the skin and outlines the structural and functional differences in adult and fetal wound healing phenotypes. A review of current developments made towards applying this clinical knowledge to promote scarless healing in adult wounds is addressed.

5 October 2017 in JCI Insight
Verified
Journal Article
Open access
PDF file
MHMichael S. Hu
Michael S. Hu
GWGraham G. Walmsley
Graham G. Walmsley
LBLeandra A. Barnes
Leandra A. Barnes
KWKipp Weiskopf
Kipp Weiskopf
RRRobert C. Rennert
Robert C. Rennert
Michael T. LongakerMichael T. Longaker
Michael T. Longaker
The monocyte lineage is essential to normal wound healing. Macrophage inhibition or knockout in mice results in impaired wound healing through reduced neovascularization, granulation tissue formation, and reepithelialization. Numerous studies have either depleted macrophages or reduced their activity in the context of wound healing. Here, we demonstrate that by increasing the number of macrophages or monocytes in the wound site above physiologic levels via pullulan-collagen composite dermal hydrogel scaffold delivery, the rate of wound healing can be significantly accelerated in both wild-type and diabetic mice, with no adverse effect on the quality of repair. Macrophages transplanted onto wounds differentiate into M1 and M2 phenotypes of different proportions at various time points, ultimately increasing angiogenesis. Given that monocytes can be readily isolated from peripheral blood without in vitro manipulation, these findings hold promise for translational medicine aimed at accelerating wound healing across a broad spectrum of diseases.

1 November 2019 in Cell Reports
Verified
Journal Article
Open access
JHJustin H. Hwang
Justin H. Hwang
JSJi-Heui Seo
Ji-Heui Seo
MBMichael L. Beshiri
Michael L. Beshiri
SWStephanie Wankowicz
Stephanie Wankowicz
DLDavid Liu
David Liu
WHWilliam C. Hahn
William C. Hahn
Androgen-receptor (AR) inhibitors, including enzalutamide, are used for treatment of all metastatic castration-resistant prostate cancers (mCRPCs). However, some patients develop resistance or never respond. We find that the transcription factor CREB5 confers enzalutamide resistance in an open reading frame (ORF) expression screen and in tumor xenografts. CREB5 overexpression is essential for an enzalutamide-resistant patient-derived organoid. In AR-expressing prostate cancer cells, CREB5 interactions enhance AR activity at a subset of promoters and enhancers upon enzalutamide treatment, including MYC and genes involved in the cell cycle. In mCRPC, we found recurrent amplification and overexpression of CREB5. Our observations identify CREB5 as one mechanism that drives resistance to AR antagonists in prostate cancers.

19 January 2023 in npj Precision Oncology
Verified
Journal Article
Open access
PDF file
ACAlexander T. M. Cheung
Alexander T. M. Cheung
EPElina L. Palapattu
Elina L. Palapattu
IPIsabella R. Pompa
Isabella R. Pompa
CAChristopher M. Aldrighetti
Christopher M. Aldrighetti
ANAndrzej Niemierko
Andrzej Niemierko
SKSophia C. Kamran
Sophia C. Kamran
Biorepositories enable precision oncology research by sharing clinically annotated genomic data, but it remains unknown whether these data registries reflect the true distribution of cancers in racial and ethnic minorities. Our analysis of Project Genomics Evidence Neoplasia Information Exchange (GENIE), a real-world cancer data registry designed to accelerate precision oncology discovery, indicates that minorities do not have sufficient representation, which may impact the validity of studies directly comparing mutational profiles between racial/ethnic groups and limit generalizability of biomarker discoveries to all populations.