Development of Treatment System by Machine Learning

Project Title

Development of Treatment System by Machine Learning

Project Area of Specialization Artificial IntelligenceProject Summary

Brain disorders are affected by chemical imbalance in the brain. These disorders arise as the person ages and may get severe to unacceptable points. Such conditions are treated by medicine as well as electrical therapies on brain. There are alternative ways to fix conditions in the world non-invasively instead of being invasive, one of them is transcranial electric stimulation (TES). It is done by placing electrodes on the patient’s scalp and providing current (in milliampere) through to the affected brain area or part. Nowadays, before providing the actual stimulation on the brain, stimulation has been simulated to check for the perfect placement of the electrode and needed current required for the dose in milliampere. Our project is to design and develop a transcranial electrical stimulation system based for the patients of common brain disorders (such as Parkinson’s). This system will be less time-consuming than conventional approaches to produce more simulated TES stimulation.

Project Objectives

The objective can be stated as:
“To design and develop a complete system for the deliverance of TES in brain

disorder patients.”

Project Implementation Method

We are developing a TES treatment system which will be personalized for different individuals. It will suggest electrode placement for every individual according to their affected area of the brain. Also, it will prescribe the sessions of TES that how much current dosage should one receive and for how long. The tools and technology being used to developed the project are python programing language using VS code IDE and core python libraries will be used for machine learning purposes for prediction algorithms and such working implementation.

We have to prepare to datasets for our project. One is for electrode placement and other is voltage dosage and treatment time sessions required for a patient.

• The electrode dataset is prepared with the help of an existing TES simulator (ROAST) providing it a number of MRIs for deep brain targeting.
• Second dataset will be prepared from existing researches and literature available.

Benefits of the Project

There are multiple software solutions for the invasive procedure and many for
non-invasive techniques but they are too costly. In this pandemic, we have
learned many lessons including one of remote management (for portability).
We aim to design a flow that will consume a patient-specific MRI and
generate a model or atlas for targeting, with (numerical and) predictive
electrode placement and predicting the treatment/prescription or dosage
required based on strategies of machine learning.
Secondly, there are a lot of TES treatment systems or software out there in the
market depending on the user requirements. Some of them provide electrode
targeting options, some of them don’t but there is a basic problem in all of
them which is time consumption. ROAST Parra lab which is a simulation
software for TES treatment takes one day to generate leadfield which is
necessary for electrode targeting. Our secondary goal is to speed up the
targeting process by using previous targeting results and then generate new
trained results with ML.

Technical Details of Final Deliverable

1. Hardware Interfaces

A 2GHz or faster processor, at least 4GB of RAM, is recommended. System is highly CPU and memory intensive (and moderately disk intensive), so concentrate on boosting those performance aspects (more memory is better, 8GB is highly recommended).

       2.Software Interfaces

Following are the software used for the tES Treatment System:

Software Used	Description
Operating System	Windows
Database	MySQL
Platform for Development	Vs Code, MATLAB
Programming Languages	Python, Flask.

Software Used

Operating System

Database

Platform for Development

Programming Languages

Final Deliverable of the Project Software SystemCore Industry HealthOther Industries Medical Core Technology Artificial Intelligence(AI)Other TechnologiesSustainable Development Goals Good Health and Well-Being for PeopleRequired Resources

Elapsed time in (days or weeks or month or quarter) since start of the project	Milestone	Deliverable
Month 1	Data Collection	2 Months
Month 2	Data Analysis	1 Month
Month 3	Electrode Prediction through ML	1 Month
Month 4	Software Building	2 Months
Month 5	Treatment Plan Data Collection	1 Month