There is a peculiar pleasure in diagnosing a condition out of your specialty: something that you studied a long time ago in your undergrad. Kind of makes up for all the unnecessary and useless stuff that you had to study back then: a partial vindication Don’t you think?

Satralizumab Phase 3 METEOROID Trial is the First-Ever Positive Trial in MOGAD (AAN 2026 / Life Science Daily News / NeurologyLive 2026) Impact: STZ ↘️68% relapse risk reduction, 79% reduction in active MRI lesions, 73% reduction in rescue therapy need. MOGAD previously had NO approved therapies and this is good news as pts relapse on existing therapies

Just went through a neuropsychological assesment of a child who was referred for difficulty with studies and behavioural phone addiction . The psychologist spent 10 minutes talking with him and sent me a report that he seemed like a bright child and did not seem to have ADHD. Contrast this with an assesment performed for his sibling in the US who underwent a detailed assesment and was finally diagnosed with mild ADHD. After starting medications, the child improved significantly. These were the tests the second child underwent 1. a clinical Interview 2.Behavioral Observations 3. Wechsler Intelligence Scale for Children - Fifth Edition (WISC-V) 4.The Beery-Buktenica Developmental Test of Visual-Motor Integration - Sixth Edition 5.Delis-Kaplan Executive Function System (D-KEFS) 6.Conners' Continuous Auditory Test of Attention 7.Conners' Continuous Performance Test - Third Edition 8.Behavior Assessment System for Children - Third Edition, Parent, Teacher and Self report forms 9.Behavior Rating Inventory of Executive Function - Second Edition, Parent and Self report forms The difference in quality of the tests was astounding and eye-opening. I was unable to even face the parent with such a shoddy test report. We need to do better than just make diagnoses based on a hunch. It's a big problem in India, where we often take shortcuts; maybe due to lack of time or large OPDs....but we compromise tremendously on quality and hardly ever make any edge contributions

CT

Through 25 years of medicine, i followed a very conventional path as did all my peers. Study a theoretical topic intensely for a few months to years, then take baby steps towards practical applications. In the case of Neurointervention, it literally took years and learning from numerous mentors and proctors, before i developed a semblance of confidence in most procedures. Coding was an esoteric and complex task that seemed out of bounds and in the realm of dedicated software developers. This made translation of my medical ideas into practical applications, almost impossible without a great deal of coordination with a developer. In the last 6-8 months, the roll out of multiple LLMs and agents has finally bridged that gap. 80-90% of the work is now planned and executed by me. The last 10-20% (app store launches/ backend work) still requires a developer right now, but it's fantastic to see my work coming to fruition.

Spring and fall: the eternal cycle of the human condition

The ESOC 2026 just concluded (May 6–8, Maastricht) with major stroke trial results : 3 positive outcomes and 2 negatives 🚀ATTRACTION (tirofiban post-EVT positive), 🚀ATILA (tirofiban in tandem lesions positive), 🚀GALLOP-2 (semaglutide neuroprotection positive), ↘️TECNO (IA tenecteplase negative), ↘️GREEN (glenzocimab futility)