Idea Name: |
ChestX Summarizer |
Slogan: |
Diagnosing made easy |
Supervisor Name: |
Prof. Dr. Muhammad Usman Akram |
Supervisor Designation: |
Professor |
Supervisor School: |
CEME |
Supervisor Department: |
Department of Computer Engineering |
Contact number: |
03336913921 |
Email ID: |
[email protected] |
Abstract: |
Radiologists are required to analyze Chest X-rays for different diseases but radiologist to patient ratio is quite low in Pakistan. ChestX Summarizer utilizes vision transformers to make a diagnosis from the available chest x-ray after which Natural Language Processing is used to provide a summary along with a detailed radiology report of the prediction that is made. |
What is the unmet need in society that your idea will fulfill ? |
ChestX Summarizer will allow radiologists to save time, and resources and attend to more patients per day. The unmet need of starting early treatment will be fulfilled due to the faster-diagnosing process which will eventually lead to a higher chance of recovery for the patient. |
Who needs it ? How many would benefit ? |
It is needed in the medical field and will especially aid the radiologists in decreasing their workload allowing them to use their resources in other works as well. Its use will also generate revenue in the medical field since all radiologists regardless of their location will have access to it. |
How will the solution works |
State-of-the-art technology like vision transformers to be utilized to create a cloud-based application that will take use of a defined algorithm to diagnose the disease from chest x-rays provided by the radiologist using trained models. It will also use Natural Language Processing to provide a summary and a detailed report to the radiologist derived from the prediction made by the model. |
Who are your competitors ? How is your solution different |
The competitors of ChestX Summarizer are SLOSH and XYLEXA. The field of work is same, but the uniqueness is the generation of a medical report along with summary of the diagnosis. The classification using vision transformers is the selling point since it’s leading-edge technology rarely used before. |
Status: |
new |
Entry Date & Time: |
2022-12-15 (0858) |