Identification of herniated disc in lumbo sacral region through MRI scans
Our goal is to facilitate the radiologist to make proper and accurate examination of MRI scans. It will reduce the time and cost because system can give results in less time with less cost. We will use Image processing techniques and Machine learning algorithms to make our system efficient enough to
2025-06-28 16:33:01 - Adil Khan
Identification of herniated disc in lumbo sacral region through MRI scans
Project Area of Specialization Artificial IntelligenceProject SummaryOur goal is to facilitate the radiologist to make proper and accurate examination of MRI scans. It will reduce the time and cost because system can give results in less time with less cost. We will use Image processing techniques and Machine learning algorithms to make our system efficient enough to identify Herniated Disc in Lumbo-Sacral region .It will be very helpful in rural areas because there is no proper facilitation for patients’, this is the major problem which we will be going to solve.
Project ObjectivesAnalyze MRI scan of lumbo-sacral region
Use these images to predict Disc Herniated in the Lumbo-Sacral region.
The objectives of the project are as follows.
- To reduce time in examination of the report.
- Once its trained, it can use professionally by radiologists to serve as an assistant.
- The machine will be trained with the authentic data sets for as much as possible accuracy.
- If there is no radiologist available, especially at Interior/Rural areas. This machine can assist and help people to recognize the problem at instance.
- To make the radiologist works automate.
Initially we have a MRI scan of backbone as an input then we will apply Image Processing techniques and operation to convert an image into digital form in order to get an enhanced image.
We will convert the input image into gray scale then we apply segmentation to break image into constitute parts or objects. It will help to extract the desired information then we will compare the images for recognizing an Lumbo-Sacral region. If image is not recognized we will apply the methods of visualizations to observe the objects clearly. As we discussed above, our model predicts the result through the disturbance of color and gap between the spines so we will apply the measurement techniques to measure the gaps between the spines.
We will apply ‘Logistic Regression Algorithm’ for predicting the results ,first we will take an image as a testing data and compute the regression coefficients of training data by the sigmoid function then we will find the relationship between the training data and the testing data and finally the system predicts the result as a output.
Benefits of the Project- To reduce time in examination of the report.
- Once its trained, it can use professionally by radiologists to serve as an assistant.
- The machine will be trained with the authentic data sets for as much as possible accuracy.
- If there is no radiologist available, especially at Interior/Rural areas. This machine can assist and help people to recognize the problem at instance.
- To make the radiologist works automate.
In the end we would deliver a module of Raspberry pie attached with GPU CARD, 16-GB Memory card and touchscreen LCD. Our module would be trained with 10000+ data sets. The image would be uploaded by external storage where the MRI of the patient will be compared to a normal person MRI. Our basic goal is to check if the patients' is affected with Disc Herniated in Lumbo-Sacral region or not. It would be done by regression and matched/compared through Image processing and Machine-Learning Techniques. The result will be shown on the LCD attached.
Final Deliverable of the Project Software SystemCore Industry ITOther Industries Medical Core Technology Artificial Intelligence(AI)Other Technologies Augmented & Virtual RealitySustainable Development Goals Good Health and Well-Being for PeopleRequired Resources| Item Name | Type | No. of Units | Per Unit Cost (in Rs) | Total (in Rs) |
|---|---|---|---|---|
| Total in (Rs) | 17000 | |||
| Power Supply | Equipment | 1 | 2500 | 2500 |
| RAM | Equipment | 4 | 1500 | 6000 |
| Hard Disk | Equipment | 1 | 2000 | 2000 |
| processor 3.2 | Equipment | 1 | 4000 | 4000 |
| casing | Equipment | 1 | 1500 | 1500 |
| cables | Miscellaneous | 5 | 200 | 1000 |