Rajan Kashyap

@bttyeo @tianchuzeng @kkli20111 @ZShaoshi @ten_photos Any guesses for appropriate data size needed for SHARP

In a meta-analysis of 210 biomedical AI studies that statistically compared models under cross-validation, 97% used invalid statistical tests. Here's our new preprint doi.org/10.64898/2026.… led by @tianchuzeng @kkli20111 @ZShaoshi @ten_photos 1/N

Eye opener 👀

Excited to put the cherry on the top for our multi-organ, multi-omics biological aging clock work: After developing 3 brain MRI-based clocks (MRIBAG), 9 phenotype-based clocks (PhenoBAG), 11 proteome-based clocks (ProtBAG), 5 metabolome-based clocks (MetBAG), we have now published 7 MRIBAGs (nature.com/articles/s4159…) from the brain and other body systems using multi-organ MRI data. Great teamwork from @chrisdav66 at Penn, Luigi Ferrucci & Keenan Walker (BLSA) from @NIH and @NIHFunding, @AndrewZalesky from University of Melbourne, Michael Raffi # Paul Aisen (A4) from USC, and many others. 1/6🚩

Dear @AlzheimersRes team- From Last 3 months there is constant harassment regarding APC payment. It is pity to see that such a big publication house has such poor management and there is constant mistakes from your end and no action to the 25 emails sent . Kindly help.

Our latest study with @SagarikaBhatta & co from @NIMHANS_BLR @NTUsg @JohnHopkinsUn . It is possibly the #first personalised #tdcs study and shows the benefits of converting non-responders to responders of treatment. Era of #precisionmedicine starts with funding from @DBTIndia

Our new work in @AlzheimersRes "How the network topology in Variants PPA patients changes over time rdcu.be/eBs2j . The work shows common and variant-specific unique networks. Thanks to @DBTIndia for funding @NIMHANS_BLR ,& @NTUsg @hkbaptistu @hopkinsneurons

🚨 New Preprint 🚨 Targeting intracranial electrical stimulation (ES) to network regions defined within individuals causes network-level effects By Cyr et al. Q: Can we use individualized network maps from precision fMRI to modulate a targeted network via ES? A: Yes! 🧵:

We also evaluated the #rsfmri & regional atrophy of each variant using CCA. We find Frontal #tdcs montage optimal only for non- fluent #PPA and dorsal for semantic #PPA rehabilitation. Thanks to Collab @cblntu @NTUsg @k_Tsapkini @JohnHopkinsUn supp 4m @USCLONI @FTDPIA

Here we used T1 weighted images of patients with semantic #PPA and non-fluent #PPA simulated three #tdcs montages- Frontal,Dorsal&Ventral using I-SATA Tool developed by us previously doi.org/10.21979/N9/KW…

All variants of Primary Progress Aphasia #PPA receive same #tdcs frontal montage treatment n report inefficacy. Our study rdcu.be/ewgi6 in @AlzheimersRes ledby @SagarikaBhatta @NIMHANS_BLR investigated the variant specific TDCS montage for #PPA rehab @DBTIndia 💲💸work

1/11 Excited to share our @Naturestudy led by @Leon_Oo1 @csabaorban @ZShaoshi doi.org/10.1038/s41586… It is well-known that AI performance scales with logarithm of sample size (Kaplan, McCandlish 2020), but in many domains, sample size can be # participants or # measurements...

@_KarishmaK Thanks for pointing truth.

It took me a while to articulate why the Ramalingaswami Fellowship does not get the prestige it deserves. 🙏 Whenever I have explained the fellowship to international colleagues, they are wide-eyed with awe. 'I wish my country had such a program' is the typical response. 🌎 It is because - in India, the prestige of permanent positions outweighs the scientist, and therefore fellowships are undervalued.

Many cool results from this paper, including a multivariate twist of Simpson's paradox! Stronger salience FC at baseline predicts better cognition at baseline. Would children with larger salience FC increase also enjoy greater cognitive gains over time? Or is the reverse true??

Prioritising Quality of education

While the world burns, we cook up a new preprint! doi.org/10.1101/2025.0… Biophysical modeling is a key tool to derive mechanistic insights into the brain. These models are governed by biologically meaningful parameters (unlike deep neural nets), but the dirty secret ... 1/N

@arpansview He has a mesmerising voice... especially in this song his voice served justice to ot

One of the best singers of Bengal is no more! RIP Pratul Mukhopadhyay youtu.be/u5zkgaOuQ8A?si…

1/ New Starstim study alert in MDD! 🧠⚡ A randomized controlled trial compared optimized multichannel tDCS vs conventional tDCS for treating major depressive disorder (MDD). Findings? Optimized tDCS outperformed conventional and sham tDCS in reducing depressive symptoms and improving cognition. 2/ The study involved 60 MD patients receiving 30 sessions of either computationally-optimized multichannel tDCS (7 electrodes, 4.0 mA), conventional tDCS (2 electrodes, 2.0 mA), or sham tDCS. Both active protocols reduced symptoms, but optimized tDCS showed faster and greater improvements. 3/ Key results: Optimized multichannel tDCS was clinically superior to both conventional and sham tDCS. Optimized tDCS reduced depressive symptoms as early as the 10th session, with a 73% response rate at 1 month (Hedge’s g=2.8). Conventional tDCS showed (milder) benefits only after the 30th session. Both outperformed sham tDCS. 4/ Cognitive improvements were also significant with optimized tDCS: better working memory, sustained attention, and executive functions compared to conventional and sham tDCS. This suggests tDCS can target both mood and cognitive deficits in MD. 5/ Brain connectivity improved, too. Optimized tDCS enhanced functional connectivity in key networks (e.g., prefrontal-temporal regions), which are often disrupted in depression. This aligns with symptom reduction and cognitive improvements. 6/ Why does this matter? Computationally optimized multichannel tDCS offers a more targeted, effective, and faster-acting treatment for MD compared to conventional tDCS. tDCS is also non-invasive, affordable and can be used at home, and has fewer side effects than medications. 7/ Bottom line: Optimized multichannel tDCS is a promising intervention for MD, with benefits for both mood and cognition. Future research should explore its potential for other neuropsychiatric disorders. 8/ Efficacy results are in line with our earlier open label home study: frontiersin.org/journals/psych… #telemedicine #hometdcs 8/ Link to paper: medrxiv.org/content/10.110… #OpenAccess #ScienceTwitter @neuroelectrics Special thanks to Ali and Michael and the team for all the heavy lifting. #starstim #stimweaver #tDCS #mdd #depression #neuromodulation #MentalHealth #DepressionTreatment #BrainStimulation #Neuroscience #BrainHealth #MentalHealthTech #Innovation

@kris_sg @CBR_IISc @iitmadras @NCBS_Bangalore @iitgn @IndiaDST Sir...invest in my lab at #NIMHANS...that aims at precision based Non-invasive treatment of neuropsychiatric disorders and be a part of the change future of neuroscience is expecting

Invested in supporting brain research at @CBR_IISc and SGBC @iitmadras. Support research at @NCBS_Bangalore @iitgn. Spend time with NMICPS @IndiaDST. Support some #deeptech startups.

For neurological disorders with atrophy, conventional is more safer than high definition #tdcs for personalised dose requirement #dementia

Excited to share the study led by @SagarikaBhatta n teams at @NIMHANS_BLR @NTUsg @JohnsHopkins n @DBTIndia showing how to select current dose in #tdcs of each patient with and without atrophy. #Precisionmedicine (1/3) mdpi.com/2076-3425/14/1…

For psychiatric disorders without cortical atrophy, both conventional and Hdtdcs can be used for dose controlled Tdcs #depression (2/3)