A-CADMCI

Computer-Aided Diagnosis Tool for Mild Cognitive Impairment based on Cognitive Tests and Decision Trees.

Motivation

A-CADMCI is built on top of scikit-learn. It was developed to aid in the diagnosis of Mild Cognitive Impairment by analyzing the scores of different cognitive tests:

• Mini–Mental State Examination (MMSE)
• Montreal Cognitive Assessment (MoCA)
• Clinical Dementia Rating (CDR)
• Global Deterioration Scale (GDS)

A-CADMCI contains decision tree classifiers trained with different databases like ADNI to predict the level of patient's cognitive impairment and guide the clinicians in the diagnosis of the disease.

Dependencies

A-CADMCI is tested to work under Python 3.4 The required dependencies to run the software are:

• Scikit-learn 0.17.1
• PyQt5 5.5.1
• Numpy 1.11.1
• SciPy 0.18.0
• Working C/C++ compiler

For easy installation on Windows, download and install WinPython 3.4.4.4Qt5.

Run the code

To run the code open a console in /src and use:

python main.py

File Structure for Model Training and Validation

Models are already trained and validated but you can also perform those tasks if you want.

For ADNI Database

A CSV file containing the following columns:

• CDR: {0, 0.5, 1, 2, 3}
• MMSE: {0, 1, 2,..., 30}
• Age: {50, 51, 52,..., 100}
• Education: {0, 1, 2,..., 20}
• Diagnosis: {Mild, Sane, Serious}

Example of the CSV file for ADNI database:
CDR,MMSE,Age,Education,Diagnosis
0,30,79,16,Sane
0,30,89,16,Sane
0.5,29,81,16,Mild
0.5,26,70,20,Serious
0.5,25,69,14,Mild

For Caldas Database

A CSV file containing the following columns:

• CDR: {0, 0.5, 1, 2, 3}
• MMSE: {0, 1, 2,..., 30}
• MoCA: {0, 1, 2,..., 30}
• GDS: {1, 2, 3,..., 7}
• Diagnosis: {Mild, Sane, Serious}

Example of the CSV file for Caldas database:
CDR,MMSE,MoCA,GDS,Diagnosis
0,30,25,1,Sane
0.5,23,14,3,Mild
0,27,15,2,Sane
2,13,4,4,Serious
0,30,24,1,Sane