ansh tandon (@anshtdn) 's Twitter Profile
ansh tandon

@anshtdn

hospital ops @palantirtech; building corpusanalytix; bioe + datasci @penn

ID: 1797994540042727424

linkhttp://corpusanalytix.com calendar_today04-06-2024 14:11:35

70 Tweet

58 Takipçi

170 Takip Edilen

Digi Biz Insight (@digibizinsight) 's Twitter Profile Photo

By strictly respecting privacy, even healthcare data can be monetized. The marketplace option for patients appears to be the fairest and most inclusive because of the inherent benefit to people. Thanks Antonio Grasso #DataPrivacy #Healthcare #DataMonetization

By strictly respecting privacy, even healthcare data can be monetized. The marketplace option for patients appears to be the fairest and most inclusive because of the inherent benefit to people. Thanks <a href="/antgrasso/">Antonio Grasso</a> #DataPrivacy #Healthcare #DataMonetization
ansh tandon (@anshtdn) 's Twitter Profile Photo

intrigued. is there a way of predicting cancer metastasis or do doctors reactively treat tumours as they show up? #Cancer #Health

ansh tandon (@anshtdn) 's Twitter Profile Photo

i believe we'll soon be able to upload consciousness into digital avatars, effectively gaining e-immortality. humans will live in virtual environments, preserving their thoughts, memories, and personalities beyond their biological lifespan. a natural progression from agents no?

ansh tandon (@anshtdn) 's Twitter Profile Photo

imagine if we could optimise blood glucose patches to have feedback loops so that dose delivery could be adjusted in real time based on activity intensity. connect to whoop perhaps? super useful for athletes

ansh tandon (@anshtdn) 's Twitter Profile Photo

super impactful slide from my penn cs healthcare-focused class today. we can only tap into ml applications if we first focus on addressing data limitations. i’m trying to dumb the acquisition down with corpusanalytix.com

super impactful slide from my penn cs healthcare-focused class today. we can only tap into ml applications if we first focus on addressing data limitations. i’m trying to dumb the acquisition down with corpusanalytix.com
ansh tandon (@anshtdn) 's Twitter Profile Photo

cis7000 takeaway for the week: AI in DKD dx: high AUC ≠ clinical utility. human expertise crucial for edge cases, comorbidities. clinical decision support system pitfalls: alert fatigue, automation bias, spectrum effect. rigorous external validation is imperative

cis7000 takeaway for the week:
AI in DKD dx: high AUC ≠ clinical utility. human expertise crucial for edge cases, comorbidities. clinical decision support system pitfalls: alert fatigue, automation bias, spectrum effect. rigorous external validation is imperative
ansh tandon (@anshtdn) 's Twitter Profile Photo

just read a linkedin post about building profitable biobanks and "tempus for x". at corpus, i'm building tempus for x, y, and z. not just another data dump. building the next-gen biobank for healthcare. massive patient repositories for targeted analysis. streamlined searching.

just read a linkedin post about building profitable biobanks and "tempus for x". at corpus, i'm building tempus for x, y, and z. not just another data dump. building the next-gen biobank for healthcare. massive patient repositories for targeted analysis. streamlined searching.
ansh tandon (@anshtdn) 's Twitter Profile Photo

shower musing: how do aggregators (like tempus, flatiron, lifebit, dandelion etc.) price biomedical datasets for licensing? at corpus we let suppliers retain control and dictate terms since we don’t directly procure data from clinics, but still curious…

ansh tandon (@anshtdn) 's Twitter Profile Photo

Thrilled to announce a strategic alliance between CorpusAnalytiX and Ariana Pharma to harness the power of #AI and #realworld healthcare data to accelerate drug development. arianapharma.com/corpusanalytix…

Thrilled to announce a strategic alliance between CorpusAnalytiX and Ariana Pharma to harness the power of #AI and #realworld healthcare data to accelerate drug development. 

arianapharma.com/corpusanalytix…
ansh tandon (@anshtdn) 's Twitter Profile Photo

benchmarking cell segmentation in spatial transcriptomics is so dogshit. we’re still relying on pixel-level accuracy metrics like IoU and NOT shifting the evaluation framework towards biological meaningfulness. for starters why don’t we use silhouette scores on UMAPs?