Intelligent Skin Doctor
smart skin doctor application that connects patients to doctors over the internet, and they never have to wait in hospital queues. Without doctor intervention, the system uses deep learning algorithms to auto-prescribe some common viral and infectious diseases. Patients enter symptoms and upload
2025-06-28 16:33:19 - Adil Khan
Intelligent Skin Doctor
Project Area of Specialization Artificial IntelligenceProject Summarysmart skin doctor application that connects patients to doctors over the internet, and they never have to wait in hospital queues. Without doctor intervention, the system uses deep learning algorithms to auto-prescribe some common viral and infectious diseases. Patients enter symptoms and upload pictures of the affected skin area on smart app and doctors get this data on their admin panels. They send curing process of disease to patients or make an appointment in case of emergency. The hospital maintains a log of data on its website to overlook doctor-patient interaction. Doctors can sell their services through this application against a fee.
Project ObjectivesØIn the modern world of Science and Technology we are curious to know about E-technologies like E-mail & E-commerce
ØThe prototype brings the new concept of E-appointment
ØA smart skin doctor application connects patients to doctors over the internet
ØThe system uses deep learning algorithms to auto-prescribe some common viral and infectious diseases
ØIt send curing process of disease to patients with medicine recommendations or make an appointment in case of emergency
Project Implementation MethodØ
ØThe data of patients are saved on the website of a hospital which is overlooking the interaction between doctors and patients
Ø
ØThere are deep learning algorithms used on the doctor's side which auto prescribe medication to patients in case of common infectious and viral diseases
Ø
ØRaspberry Pi is used as the main server to host Tensor flow for poets 2 i.e. a deep learning library
Ø
ØThe doctor application is made on Android Studio and linked with Raspberry Pi
Ø
ØDeep learning model is trained with Acne, Melanoma, Eczema and Cystic image datasets
Ø
ØGoogle Firebase and web database are maintained
Ø
ALGORITHM PLANNING
•The system uses Tensor flow numerical computation library by Google as it has strong support for machine learning and deep learning
CONVOLUTIONAL NEURAL NETWORK (CNN)
•It is a learning model like artificial neural networks which is inspired by the neural network of human brain
•It is able to make intelligent guesses with high accuracy with the help of historical data
•Training the new set of images requires greater computing power, so we used pre-trained pieces of models
•
•The image feature extraction module is used with MobileNet V1 or V2 to train our set of images
•
•Feature vector is 1-D Tensor for representing images for classification
•
•we have gathered the images of four diseases and trained it for classification
•
•In a very lesser time of 25-35 minutes we have train our data on CPU without using expensive GPU support
Benefits of the ProjectØ An adequate solution for reducing the long ques in hospitals was required.
Ø
Ø This prototype the machine learning algorithms to train doctor experiences.
Ø This will be the time-saving & adequate solution for hospitals
Ø The accurate results ensures patient safety
Ø All the database is maintained in case of any incident
The project is portable and can be installed anywhere in local & remote areas
Technical Details of Final DeliverableCONCLUSION
•
• This paper presents a smart skin doctor application that connects patients to doctors over the internet, and they never have to wait in hospital queues
•
•
•Machine learning algorithms have been used to make this system smart
•
•
•A number of features have been integrated with this system in order to make it convenient for common people
•
•
Final Deliverable of the Project Hardware SystemType of Industry IT , Medical Technologies Artificial Intelligence(AI), Internet of Things (IoT), Cloud InfrastructureSustainable Development Goals Industry, Innovation and InfrastructureRequired Resources| Item Name | Type | No. of Units | Per Unit Cost (in Rs) | Total (in Rs) |
|---|---|---|---|---|
| Total in (Rs) | 60000 | |||
| Raspery pi kit | Equipment | 1 | 20000 | 20000 |
| Andriod Cell Phone | Equipment | 1 | 20000 | 20000 |
| Hospital Surveys Doctor Fees | Miscellaneous | 1 | 10000 | 10000 |
| Domain + hosting Server | Equipment | 1 | 10000 | 10000 |