Оффлайн
- Регистрация
- 1 Мар 2015
- Сообщения
- 1,481
- Баллы
- 155
Prepared by: Alireza Minagar, MD, MBA, MS
software engineer
Intro:
As a neurologist-turned-software engineer, I’ve explored how AI models trained on radiological, genomic, and clinical data are transforming the landscape of neurology.
Let’s break down:
? Use Case 1: Predicting Multiple Sclerosis Progression
Dataset: Longitudinal MRI scans + CSF profiles
Model: LSTM + random forest hybrid
Output: 5-year EDSS disability trajectory
? Use Case 2: Genomic Pattern Mining in Alzheimer’s
Tools: scikit-learn, BioPython, pandas
Technique: SNP frequency clustering using unsupervised learning
Result: Highlighted APOE4 and potential novel modifiers
? Use Case 3: NLP in Neurology Notes
Stack: spaCy + SciKit-learn
Task: Flagging seizure risk factors from text
Result: 78% recall, 65% precision on clinical validation
Conclusion:
AI isn’t replacing neurologists—it’s becoming their algorithmic co-pilot.
Would love to hear from others building ML models in medicine. Let’s collaborate.
software engineer
Intro:
As a neurologist-turned-software engineer, I’ve explored how AI models trained on radiological, genomic, and clinical data are transforming the landscape of neurology.
Let’s break down:
? Use Case 1: Predicting Multiple Sclerosis Progression
Dataset: Longitudinal MRI scans + CSF profiles
Model: LSTM + random forest hybrid
Output: 5-year EDSS disability trajectory
? Use Case 2: Genomic Pattern Mining in Alzheimer’s
Tools: scikit-learn, BioPython, pandas
Technique: SNP frequency clustering using unsupervised learning
Result: Highlighted APOE4 and potential novel modifiers
? Use Case 3: NLP in Neurology Notes
Stack: spaCy + SciKit-learn
Task: Flagging seizure risk factors from text
Result: 78% recall, 65% precision on clinical validation
Conclusion:
AI isn’t replacing neurologists—it’s becoming their algorithmic co-pilot.
Would love to hear from others building ML models in medicine. Let’s collaborate.