Hacker News Reader

Comment by thinkcomp | Hacker News Reader

That's one possible conclusion. More broadly, the morphology of the cerebellum matters a great deal, and minute differences from person to person A) could explain why autism appears to have such a broad array of symptoms and severities; B) have been largely ignored by medicine for decades due to mistaken convention, deep confusion among physicians over medical terminology (mega cisterna magna versus DWV) and a tendency for doctors to throw their hands in the air the minute they hear "autism" because it's simply accepted that it's too complex to diagnose; and C) suggest that brain MRIs may be the most cost-effective cost-preventive diagnostic tool available for those with autism diagnoses--which doctors hardly ever recommend due to (B).

View on HN

Why am I getting down voted? I'm just trying to understand the article. If you disagree with the article's conclusion, its the author's hypothesis not mine. He even cites cases of children whose cysts were surgically treated and now lead normal lives.

These were certainly important points of the article. However, it was a shotgun blast that covered much more, incompetent medical profession, inconsistent terminology causing confusion, stonewalls for patients diagnosed with autism, etc. it tried to capture too much in too little space.

I would think that someone would work to categorize variance in the physical and personality affects of autism... much like Myers-Briggs for personality in general. It seems to me that it should have been broken down into sub-categories (not as the article suggests in terms of cyst affect, but in terms of autism in general... they call it a spectrum, but really what is that spectrum and how is it defined. If different characteristic behaviors are given, why not just box them up in an N-by-N grid (max 4 options over 4 behaviors), which would help categorize.

I also feel that having some MRI development scans would definitely help with correlation of data here. No, not every case would be treated, but it certainly could get better in terms of diagnosis and being able to make predictions.

Imagine if all children were scanned at 6mo, 1yo and every other year until 18... that would be much more valuable data to humanity than some other uses of big data storage out there. In this case removing the cyst early might have helped... I don't think that would be the case most of the time... but by not having a better picture, it doesn't help.

We're in a first-world nation and outspend the rest of the world combined in our millitary... we really couldn't get a couple more MRI machines per hospital to handle the capacity? They don't all need to be high resolution scans to point to more information either.

I agree with the comments of many that this confuses and dillutes a few points... that said, it's not a medical paper, it's a blog article pointing out frustration in a lack of diagnostics that could help a lot of people.

> Imagine if all children were scanned at 6mo, 1yo and every other year until 18... that would be much more valuable data to humanity than some other uses of big data storage out there.

Imagine that instead, upon birth, all children had their entire genome sequenced and recorded. That would be a far greater use of big data storage.

Full genome sequencing costs a third of an MRI.

Reply on news.ycombinator.com