Skip to content

This project investigates the potential of deep learning to automatically identify and classify brain tumors from medical images. This could lead to faster and more accurate diagnoses, potentially improving patient outcomes.

Notifications You must be signed in to change notification settings

Sh-bharat/Machine_Learning_for_Medical_Image_Analysis-Brain_Tumor_Classification

Repository files navigation


Machine_Learning_for_Medical_Image_Analysis-Brain_Tumor_Classification

Overview

This project focuses on classifying brain tumor MRI images into 44 different classes using YOLOv8, a state-of-the-art object detection algorithm. The dataset used for training and evaluation is sourced from Kaggle and consists of MRI images with various types of brain tumors.

Dataset

  • Source: Brain Tumor MRI Images
  • Total Images: 44 classes with varying image counts (ranging from 31 to 43 images per class)

Classes

  1. Neurocitoma T1C+
  2. Meduloblastoma T1C+
  3. Ependimoma T2
  4. Astrocitoma T1
  5. _NORMAL T2
  6. Ganglioglioma T2
  7. Granuloma T1
  8. Carcinoma T1C+
  9. Neurocitoma T2
  10. Papiloma T1
  11. Tuberculoma T1C+
  12. Glioblastoma T2
  13. Carcinoma T2
  14. Meningioma T1C+
  15. Papiloma T2
  16. Oligodendroglioma T1
  17. Granuloma T2
  18. Schwannoma T1
  19. Astrocitoma T1C+
  20. Meduloblastoma T1
  21. Germinoma T2
  22. Papiloma T1C+
  23. Glioblastoma T1C+
  24. Ependimoma T1C+
  25. Oligodendroglioma T2
  26. Carcinoma T1
  27. Ependimoma T1
  28. Neurocitoma T1
  29. Schwannoma T2
  30. Astrocitoma T2
  31. Ganglioglioma T1C+
  32. Glioblastoma T1
  33. Schwannoma T1C+
  34. Ganglioglioma T1
  35. Germinoma T1C+
  36. Meduloblastoma T2
  37. Tuberculoma T1
  38. Germinoma T1
  39. Oligodendroglioma T1C+
  40. _NORMAL T1
  41. Tuberculoma T2
  42. Granuloma T1C+
  43. Meningioma T2
  44. Meningioma T1

Sample Predicted Image

val_batch2_pred

Model Performance

  • Accuracy: 97%

Result

Usage

  1. Training: Use the provided dataset to train the YOLOv8 model.
  2. Inference: Once trained, the model can be used to classify brain tumor MRI images. Provide an image to the model, and it will predict the class of the brain tumor present.

Got it! Here's the updated section:

Libraries Required

import zipfile
import os
import shutil
from tqdm import tqdm
import cv2
import matplotlib.pyplot as plt
from ultralytics import YOLO

Credits

Created by Bharat_Sharma
for any suggestion, Queries feel free to contact bharat05122002@gmail.com


About

This project investigates the potential of deep learning to automatically identify and classify brain tumors from medical images. This could lead to faster and more accurate diagnoses, potentially improving patient outcomes.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published