Zhaozhi Li

Zhaozhi Li

Auto-generated profile, based on publicly available information

Zhaozhi Li

Massachusetts General Hospital, Boston, MA USA

Research Focus & Metrics

Academic Career Length

5Years (2022-Present)

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

Representative Publications

View All Publications
30 August 2022 in Journal of Medical Internet Research
Verified
Journal Article
Open access
Ayush NooriAyush Noori
Ayush Noori
CMColin Magdamo
Colin Magdamo
XLXiao Liu
Xiao Liu
TTTanish Tyagi
Tanish Tyagi
ZLZhaozhi Li
Zhaozhi Li
SDSudeshna Das
Sudeshna Das
Background: Electronic health records (EHRs) with large sample sizes and rich information offer great potential for dementia research, but current methods of phenotyping cognitive status are not scalable. Objective: The aim of this study was to evaluate whether natural language processing (NLP)–powered semiautomated annotation can improve the speed and interrater reliability of chart reviews for phenotyping cognitive status. Methods: In this diagnostic study, we developed and evaluated a semiautomated NLP-powered annotation tool (NAT) to facilitate phenotyping of cognitive status. Clinical experts adjudicated the cognitive status of 627 patients at Mass General Brigham (MGB) health care, using NAT or traditional chart reviews. Patient charts contained EHR data from two data sets: (1) records from January 1, 2017, to December 31, 2018, for 100 Medicare beneficiaries from the MGB Accountable Care Organization and (2) records from 2 years prior to COVID-19 diagnosis to the date of COVID-19 diagnosis for 527 MGB patients. All EHR data from the relevant period were extracted; diagnosis codes, medications, and laboratory test values were processed and summarized; clinical notes were processed through an NLP pipeline; and a web tool was developed to present an integrated view of all data. Cognitive status was rated as cognitively normal, cognitively impaired, or undetermined. Assessment time and interrater agreement of NAT compared to manual chart reviews for cognitive status phenotyping was evaluated. Results: NAT adjudication provided higher interrater agreement (Cohen κ=0.89 vs κ=0.80) and significant speed up (time difference mean 1.4, SD 1.3 minutes; P<.001; ratio median 2.2, min-max 0.4-20) over manual chart reviews. There was moderate agreement with manual chart reviews (Cohen κ=0.67). In the cases that exhibited disagreement with manual chart reviews, NAT adjudication was able to produce assessments that had broader clinical consensus due to its integrated view of highlighted relevant information and semiautomated NLP features. Conclusions: NAT adjudication improves the speed and interrater reliability for phenotyping cognitive status compared to manual chart reviews. This study underscores the potential of an NLP-based clinically adjudicated method to build large-scale dementia research cohorts from EHRs.

16 June 2023 in Alzheimer's & Dementia
Verified
Journal Article
Open access
ZLZhaozhi Li
Zhaozhi Li
AWAstrid Wachter
Astrid Wachter
SASrinija Alla
Srinija Alla
Ayush NooriAyush Noori
Ayush Noori
AAAicha Abdourahman
Aicha Abdourahman
ASAlberto Serrano-Pozo
Alberto Serrano-Pozo
Background: Microglia, astrocytes, and neurons undergo profound functional alterations in response to amyloid‐β (Aβ) plaques and neurofibrillary tangles (NFTs) in Alzheimer’s disease (AD). Recent studies using human induced pluripotent stem cell‐derived glial cells and neurons, APOE knock‐in mice, and human brain bulk RNA‐seq have implicated the APOE genotype in these changes, but lack spatial information with respect to Aβ plaques and NFTs. Here we tested the hypotheses that (1) transcriptomic differences between AD and control subjects are maximum within and near Aβ plaques and NFT‐bearing neurons; (2) Aβ plaques and NFTs are associated with distinct transcriptomic changes; and (3) the APOE genotype differentially impacts the transcriptomic changes associated with Aβ plaques and NFTs in the AD brain.Method: Laser capture microdissection (LCM) in cryostat sections was performed from superior temporal gyrus (BA22) cortex of human AD (n = 10, including n = 5 APOEε4/ε4, n = 4 APOEε3/ε3, and n = 1 APOEε3/ε4) and age‐ and sex‐matched controls (n = 8). Thioflavin‐S‐positive Aβ plaques, the 50 µm halo around them, NFTs with the 50 µm halo around them, and areas far (>50 µm) from plaques and NFTs were laser‐capture microdissected and subjected to RNA‐sequencing to identify differentially expressed genes (DEGs).Result: Aβ plaques had a greater impact on the transcriptome than NFTs (i.e., higher number of DEGs, both upregulated [logFC>0, unadj. P‐value<0.05: 2,623 vs. 1,230 genes; adj. P‐value0, unadj. P‐value<0.05: 2,563 vs. 1,056 genes; adj. P‐value plaque halo > NFTs > far areas was evident, with upregulation of neuroinflammation and downregulation of synaptic neurotransmission and energy metabolism‐related gene sets. Comparing APOEε4 and APOEε3 age‐ and sex‐matched homozygotes revealed greater changes in APOEε4 homozygotes across locations.Conclusion: Aβ plaques and, to a lesser extent, NFTs, concentrate the bulk of transcriptomic changes in the AD cortex. The APOEε4 allele is associated with greater microglial and neuronal transcriptomic responses to Aβ plaques and NFTs, compared to APOEε3 carriers. These findings will inform future spatial transcriptomics studies.