1. Home »
  2. Supervisor: Dr. Raheel Siddiqi
Supervisor: Dr. Raheel Siddiqi

 

Visual Recognition of Doctor’s Handwritten Prescriptions

 

Supervisor: Dr. Raheel Siddiqi, Senior Assistant Professor

Campus/School/Dept: CS Department, BUKC

Supervisory Record: 

  • PhD Produced: 0
  • PhD Enrolled: 1
  • MS/MPhil Produced: 2
  • MS/MPhil Enrolled: 0

 

Topic Brief Description: 

Doctors’ handwritten prescriptions are important documents for patients and their family as well as caregivers. It is quite common that pharmacist, patients and caregivers find it difficult to recognize doctor’s handwriting on prescriptions. This is especially true for countries where handwritten prescriptions are still the norm. This leads to errors which may result in a number of serious issues like incorrect dosage, consumption of wrong, unprescribed medicine, etc. This research aims to develop a reliable mechanism that can recognize doctors’ handwriting with a high level of accuracy. The research is primarily focused on handwritten prescriptions of doctors in Pakistan. Pakistani doctors often mix English and Urdu text in their prescriptions so the proposed research will target text of both languages.

 

Research Objectives/Deliverables: 

  1. To develop a reliable mechanism for recognition of handwritten prescriptions.
  2. To exploit deep learning and computer vision technologies to solve this problem
  3. Research  Questions
1. To develop a sufficiently large, labeled dataset of handwritten prescriptions by doctors based in Pakistan.
2. What is the most effective and optimized mechanism to recognize doctor’s handwritten prescriptions?
3. How to create a comprehensive labeled dataset for the problem?

 

Candidate’s Eligibility Profile: 

  1. The applicant must have an MS/MPhil/Equivalent degree in computer science/artificial intelligence/computer engineering with CGPA >0. Besides, applicants must have a strong background in mathematics, computer vision and deep learning.
  2. Experience in designing deep learning solutions using frameworks like TensorFlow or PyTorch is desirable. Candidates should have excellent communication and interpersonal skills.
  3. Proficiency in spoken and written English is essential.