Skip to content

takshkothari/HealthInspect

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

83 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Important!!

The new address of the website is: HealthInspect.

HealthInspect

HealthInspect is a web-based application that allows users to upload images of diseases and receive accurate, reliable information about their condition in just seconds. The uploaded image is processed through a powerful AI model, which uses deep learning algorithms and advanced image recognition techniques to analyze the image and identify any signs of disease.

Objective

The objective of HealthInspect is to provide users with a user-friendly, accessible tool for self-diagnosing medical conditions. The first product launched is the Skin Disease Detector.
With HealthInspect, users can take images of their skin diseases from the comfort of their own home using just their smartphone camera and receive accurate, reliable information about their condition in just seconds. This information is presented in a clear, easy-to-understand format that anyone can follow, making it an ideal solution 24/7 and for individuals who may not have immediate access to a healthcare provider.

Methodology and Results

HealthInspect is built on a Google Cloud Platform, which is a powerful, scalable platform for building web applications. The application uses a Vertex AI Model deployed on an endpoint, trained on the dermnet dataset stored in a Cloud Storage Bucket.

Classifying skin diseases is a complicated task in medicine due to the similarities between different diseases, but has huge scope compared to other fields of medicine.
Our model identifies 21 different classes of skin diseases with an average accuracy of 77.7%, which is one of the highest number of classes a model has been accurately trained on according to various research papers, providing a large coverage of common types.They are:

Acne or Rosacea
Actinic Keratosis Basal Cell Carcinoma or other Malignant Lesions
Atopic Dermatitis
Bullous Disease
Eczema
Hair Loss Alopecia or other Hair Diseases
Herpes HPV or other STDs
Light Diseases or Disorders of Pigmentation
Lupus or other Connective Tissue diseases
Melanoma Skin Cancer Nevi or Moles
Nail Fungus or other Nail Disease
Poison Ivy or other Contact Dermatitis
Psoriasis Lichen Planus or related diseases
Scabies Lyme Disease or other Infestations or Bites
Seborrheic Keratoses or other Benign Tumors
Systemic Disease
Tinea Ringworm Coridiasis or other Fungal Infections
Urticaria Hives
Vascular Tumors
Vasculitis
Warts Molluscum or other Viral Infections

Note: the model currently does not predict 'none of these' as of now due to lack of data.

The web application is built using Flask, a Python web framework. The application is deployed on a Compute Engine instance in Google Cloud that runs on AMD EPYC Milan platform with a T2D type machine. The web app is deployed using mod Apache2 with WSGI used to connect the Flask app to the external IP of the VM. The site is accessible via the external IP.

The front-end of the application is built using JavaScript, HTML, CSS, and Bootstrap for designing. The website also handles compression on large images to ensure quick and smooth processing.

Instructions

To use HealthInspect, simply go to the following website: HealthInspect.
On the homepage, select product, and upload an image of the affected area on our super lightweight and user-friendly platform, and thats it! The uploaded image is then processed through our powerful Vertex AI model, which is deployed on the Google Cloud platform. Once the diagnosis is displayed, our system provides a description of the symptoms, causes, and possible home remedies for the identified disease.

For more information on how to use HealthInspect, please watch: demovideo

Conclusion

We are confident that HealthInspect will continue to have a positive impact on the lives of people around the world. The application is hosted on the Google Cloud Platform, which ensures its reliability and scalability. Furthermore, our mobile-friendly and blazingly fast interface, makes it easy for anyone to use the application, regardless of their technical expertise. We had the opportunity to present this to Dr. M.S. Narula from Chandigarh, and this was what he had to say: testimony.
We believe that HealthInspect has the potential to become a valuable tool in the fight against skin diseases and more, and we are excited to continue to develop and improve and exand the application in the future.

About

No description, website, or topics provided.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Contributors 4

  •  
  •  
  •  
  •