Idea Details

Team and Contact Details

SNo Student Name School Degree Year Email
1Usama BaigSEECSUndergraduateFourthubaig.bese15seecs@seecs.edu.pk
2Muhammad Mahad KhanSEECSUndergraduateFourthmmahad.bese15seecs@seecs.edu.pk
3KhalidSEECSUndergraduateFourthkhalid.bese15seecs@seecs.edu.pk
4Haris AliSelectUndergraduateFourthharisbaig843@gmail.com
5Saad FarooqASABUndergraduateFourthusamabaig007@gmail.com

Inter School Idea ? Yes
Do you need expertises from another area: Yes
If Yes please provide details of expertises you need: Pathology Biosciences

Idea Details

Idea Name: Catanet
Slogan: Positively Pink!
Startup Domain: Healthcare
Abstract:
    A deep learning pipeline for automatic detection, localization and classification of breast cancer metastasis in whole slide images of histopathological lymph node section.
What is the unmet need in society that your idea will fulfill ?
    Slide-level analysis is an exhaustive task for pathologists which is pivotal in deciding cancer stage and eventually for treatment.If done by experts it takes ~30 hrs with only 73% accuracy.But out of every 100 Stage-1 patients,only 10 are diagnosed in Pak, reducing the survival rate from 98% to 50%
Who needs it ? How many would benefit ?
   In Pakistan, 10.2M people are vulnerable to breast cancer. Out of 80000 diagnosed, 40000 die every year due to late diagnosis. If diagnosed at Stage 1, their survival rate can be raised to above 85%.
How will the solution works
    Currently slide analysis is performed by pathologist. Our pipeline model will take this slide as input and automate the process by 1.Extracting Region of Interest(ROI). 2. Extracting +ve and -ve patches from slide. 3.Training our Deep Model on these patches. 4.Localizing tumor by probability heat maps generated by Deep Model 5.Performing slide-level classification 6.Determining the metastasis stage.
Who are your competitors ? How is your solution different
    Currently in Pakistan, this process is done manually by pathologists. We are the only one presenting the automated solution with high accuracy.
Status: new
Entry Date & Time: 2018-12-23 (1731)