Machine Learning — Can it lead to better diagnosis of Schizophrenia?
Schizophrenia:
It is a chronic, severe and a disabling mental disease. It is found all over the world and yet it is still unknown by many. Approximately 1 percent of the population develops this complex disorder.
People with schizophrenia often suffer terrifying symptoms such as hearing internal voices not heard by others, or believing that other people are reading their minds, controlling their thoughts, or plotting to harm them.
There are treatments available, but it has been estimated that the treatment has to be taken throughout their lifetime and this may cause dangerous side-effects.
It is one of the hardest illness that can be diagnosed and can have devastating long-term outcomes due to the below said reasons:
- It is often misdiagnosed with other illnesses.
- When diagnosed, it is harder to predict whether a prescribed set of drugs/treatment will help the patient recover.
- Unable to diagnose at early stages.
- Patients are generally aggressive and hard to treat.
- Lack of awareness.
HOW could Machine Learning help to solve this problem?
Though research is gradually unraveling the complex causes of the disease, the sole study of biological study is not sufficient.
Here comes the methods of imaging the brain’s structure and function which promises new insights into the disorder.
— — — — — — —
Who could have thought that ML could solve such problems? Who could have expected that disjointed domains of study could come together this soon!?
— — — — — — —
A research from the University of Alberta(U of A) uses a Machine Learning algorithm to examine fMRI images of both newly diagnosed, previously untreated schizophrenia patients and healthy subjects.
By measuring the connections of a brain region called the superior temporal cortex to other regions of the brain, the algorithm successfully identified patients with schizophrenia at 78 percent accuracy. — Better Diagnosis.
It also predicted with 82 percent accuracy whether or not a patient would respond positively to a specific antipsychotic treatment named risperidone. — Prediction for using right treatment.
“This is the first step, but ultimately we hope to find reliable biomarkers that can predict schizophrenia before the symptoms show up,” said Cao, an assistant professor of psychiatry at the U of A.
“We also want to use machine learning to optimize a patient’s treatment plan. It wouldn’t replace the doctor. In the future, with the help of machine learning, if the doctor can select the best medicine or procedure for a specific patient at the first visit, it would be a good step forward.”
Tackling such problems:
This is the era of learning, through solving problems. We have access to unlimited resources, now is the time to pick up problems and solve them. Let’s use our “diverse knowledge” and build a healthy, functional society together.
Let’s craft a better future! :)
