AI in Hospitals: How Artificial Intelligence Is Saving Lives

I'm Parikshet More, I'm 11 years old, and last year I asked my dad: if AI can beat grandmasters at chess, why can't it cure cancer? The answer taught me more about medicine than any school lesson.

The Short Answer

AI cannot cure cancer by itself — but it is already detecting certain cancers earlier than most human doctors can. That's a genuinely big deal, because catching cancer early is when treatment actually works.

Reading X-Rays and Scans

When a radiologist looks at a chest X-ray, they are pattern-matching thousands of hours of training against what they see. AI does the same thing — except it has been trained on millions of scans, never gets tired, and does not miss the 3am shift. Google Health's AI detected breast cancer in mammograms with fewer false positives than human radiologists in a 2020 study published in Nature. NHS hospitals in the UK are now using it as a second reader on every scan.

I found this exciting and a little scary at the same time. Exciting: earlier detection saves lives. Scary: what if the AI is wrong and no human double-checks?

Predicting Who Gets Sick

During COVID-19, AI models predicted which patients admitted to hospital were most likely to deteriorate in the next 24 hours. Hospitals used this to prioritise ICU beds. BlueDot, a Canadian AI company, actually flagged unusual pneumonia clusters in Wuhan nine days before the WHO issued its first COVID-19 public alert — by scanning airline ticketing data and news in 65 languages.

That is wild. An algorithm reading plane tickets noticed a pandemic before the world's top health body did.

Drug Discovery: Compressing Decades into Years

Creating a new drug normally takes 10–15 years and costs over a billion dollars. Most fail. DeepMind's AlphaFold solved the protein-folding problem in 2020 — predicting the 3D shape of any protein from its genetic code. This matters because drug molecules work by fitting into protein shapes like a key into a lock. If you know the shape, you can design the key faster. Scientists have already used AlphaFold to fast-track drugs for malaria, antibiotic-resistant infections, and rare genetic diseases.

Robotic Surgery

The Da Vinci Surgical System is a robot guided by a human surgeon sitting at a console. The robot's arms are steadier than any human hand and can make cuts smaller than a millimetre. AI assists by flagging important structures — "that's a blood vessel, don't cut there" — and learning from every procedure to improve future ones. More than 10 million surgeries have been performed with Da Vinci systems worldwide.

The Limits (And Why They Matter)

Medical AI has real problems. It sometimes performs worse on populations that were underrepresented in its training data — so an AI trained mostly on light-skinned patients may miss conditions on darker skin tones. This is a fairness problem that researchers are actively working on. There is also the question of accountability: if an AI misses a diagnosis, who is responsible — the hospital, the software company, or the doctor?

My view: AI in medicine is like having a brilliant assistant who reads fast and never gets tired, but still needs a qualified doctor in charge. The goal is not replacing doctors. It is giving every patient access to the kind of expert analysis that only big hospitals currently afford.

What This Means for You

If you want to work in medicine one day, adding AI skills to your toolkit — even just understanding how these tools work — will make you a better doctor, nurse, or researcher. The future of healthcare is human + AI, not human vs AI.