Aap blood work karwate hain. Doctor kehte hain sab normal hai. Aap ghar chale jate hain.
Jo kisi ne nahi bataya, woh ye ke un mein se ek number pichle saal se 15% zyada tha. Aur do saal pehle se 30% zyada. Woh ab bhi "normal" range mein hai, is liye na koi flag, na call, na follow-up. Lekin woh badal raha hai. Aur woh kis taraf ja raha hai, ye is se zyada ahem hai ke abhi kahan hai.
Aaj istemal hone wala har clinical system ek hi tarah kaam karta hai. Ye number line se upar hai ya neeche? Haan ya nahi. Modern healthcare ka poora infrastructure isi sawal ka jawab dene ke liye bana hai. Aur sawal hi ghalat hai.
Asal sawal
Ek hi din, do charts mein 1.4 ka creatinine aata hai.
Pehle mareez ka creatinine atharah mahine pehle 0.9 tha. Phir 1.0. Phir 1.2. Ab 1.4. Isay aage project karein to ye mareez ek saal ke andar kidney disease ki taraf ja raha hai. Lekin 1.4 alert threshold se neeche hai. Koi alarm nahi bajta. Visit "repeat in three months" par khatam ho jati hai.
Doosra mareez teen saal se 1.4 par hai. Stable. Managed. Routine.
Number wahi. Ek mareez mushkil mein hai. Doosra bilkul theek. Aaj clinical use mein koi system ye farq nahi bata sakta. Na Epic. Na Cerner. Na un ke upar bane AI tools mein se koi. Sab 1.4 dekhte hain aur ek hi sawal poochte hain: kya ye line se upar hai?
Sahi sawal ye hai: ye kahan se aaya hai, kahan ja raha hai, aur kitni tezi se?
Har physician ye janta hai. Har system isay nazar andaz karta hai.
Jab main mareez dekhta tha, to jin behtareen clinicians ke saath maine kaam kiya, woh aisa kaam karte the jo EHR un ke liye nahi kar sakta tha. Woh purane labs nikalte aur raddi kaghaz par trend lines kheenchte. Teen saal ki values dimagh mein rakhte aur direction ke baare mein sochte. Woh ye nahi poochte the ke "kya ye number normal hai?" Woh poochte the "kya ye mareez badal raha hai?"
Yahi reasoning achi clinical medicine ki buniyad hai. Aur yahi woh cheez hai jo bare paimane par chalne wala koi clinical system nahi karta. EHR billing ke liye encounters document karne ko bana tha. Alerts thresholds check karne ko bane the. AI tools unhi threshold-based systems ke upar bane. Technology ki teh par teh, aur sab se bunyadi clinical sawal be-jawab rehta hai: kya ye mareez waqt ke saath behtar ho raha hai ya bigar raha hai?
Trajectories na parhne ki qeemat
United States mein 37 million adults ko chronic kidney disease hai. Un mein se 90% ko is ka pata hi nahi.
Ye is liye nahi ke data maujood nahi. Un ka creatinine saalon se khisak raha hai, chart mein nazar aata hai, har visit par dekha jata hai, file hota hai aur bhula diya jata hai. CKD jaldi pakri jaye to saal ke $2,000 se $5,000 mein manage ho sakti hai. Der se pakri jaye to dialysis ka kharcha saal ka $90,000 hai. In dono nateejon mein farq bas itna hai ke kisi ne trajectory parhi ya nahi.
Yahi pattern diabetes, heart failure, COPD aur hypertension mein bhi dohraya jata hai. Saalon ka drift jo pakra ja sakta tha, un systems ko nazar nahi aata jo sirf ye poochte hain ke aaj ka number line se upar hai ya nahi.
Koi aur ye kyun nahi karta
Zyada tar clinical AI snapshots par train hoti hai. Ek visit. Ek lab draw. Waqt ka ek lamha. Model us lamhe ko high-risk ya low-risk mein classify karna seekhta hai. Usay aaj ka creatinine 1.4 nazar aata hai. Usay kuch andaza nahi ke atharah mahine pehle creatinine kya tha.
Snapshot se trajectory nahi parhi ja sakti. Ek point se slow drift nahi pakra ja sakta. Architecture khud is ki ijazat nahi deta. Snapshot model mein mazeed data daalne se ye theek nahi hota. Model ko bara karne se bhi nahi. Mareez ko represent karne ka tareeqa badalna hoga. Waqt ko asal dimension banana hoga, baad ki soch nahi.
Yahi cheez Chroneme ko market ki har doosri cheez se alag karti hai. Usi data par laga behtar algorithm nahi. Sehat ko represent karne ka buniyadi tor par alag tareeqa. Mareez saalon aur providers ke paar ek musalsal signal ki tarah maujood hai, ek doosre se kati hui visits ke silsile ki tarah nahi. System parhta hai ke cheezein waqt ke saath mil kar kaise badalti hain, aur pehchanta hai ke kab koi trajectory aisi jagah ja rahi hai jahan tawajjo zaroori hai.
Amal mein is ka matlab kya hai
Pehla mareez flag hota hai. Doosra nahi. Creatinine wahi. Trajectories alag. Recommendations alag.
Isi qism ki tabdeeli pakarne ke liye hum Chroneme bana rahe hain. Hum isay ek pre-registered protocol ke tehat validate kar rahe hain, aur jo kuch milega publish karenge, woh jagahen bhi jahan ye poora nahi utra.
Jo tabdeeliyan sab se pehle ahem hoti hain, woh shayad hi kabhi dramatic hoti hain. Woh ahista ahista hone wale drifts hote hain. Har quarter 0.1 barhta creatinine. Har mahine chand beats upar khisakti heart rate ki baseline. Hafton mein ahista ahista girta oxygen level. Aise patterns jinhein koi akela data point kabhi flag nahi karega. Aise patterns jin ke liye threshold-based systems apni banawat hi se andhe hain.
Clinical intelligence ki agli nasal yahin basti hai. Emergencies pakarne mein nahi. Ahista chalne wali emergencies ko tez hone se pehle pakarne mein.
Sources: CKD prevalence ke figures National Kidney Foundation se.
Naye essays aur updates, email par.