Dodging DDIs

We aim to provide potential drugs for patients with predefined set of medications. We consider drug-drug interactions to prevent adverse effects and provide alternative medications.

Comment

DRUG BREWERS

Project: DODGING DDIs

Inspiration Our team wanted to address the challenge of quick decision making in the general clinical scenario (emergency room) for assisting pharmacists.

What it does We aim to provide a list of potential drugs for a patient who comes into the clinic with a predefined set of medications for their existing conditions. In order to assist the clinical decision making, we will factor in drug-drug interactions to prevent polypharmacy adverse effects and provide a list of alternative medications that can be prescribed along with their pharmacokinetic data.

How we built it Our team used two datasets from the Stanford Biomedical Network Dataset Collection (BioSNAP) repository: Interactions between FDA approved drugs, Side effects of Drug combinations. For the drug-to-ID mapping, we used a dataset from the DrugBank. Using Python, we tried to map the disease to the potential drugs that can be given by eliminating those interacting with the medications already taken by the patient.

Challenges we ran into This was our first hackathon, and we had great trouble coming up with a workable idea to implement within the limited time period. The biggest challenge was finding the right data to be able to simulate the problem at hand. Especially considering the healthcare scenario, clinical data is not easily accessible. We had to alter our project aim to fit the data that was available online, which was a huge challenge.

Accomplishments that I'm proud of Brainstorming and coming up with ideas that have huge potential in the medical industry. Learning about pharmacology being from a background of research and tech. Working together and bonding as a team

What I learned

  • Pharmacology concepts
  • Importance of medication management
  • Insight into US Healthcare system
  • EHR - format, extracting relevant information
  • Extracting information from open source datasets

What's next for Dodging DDIs

  • Generate possible set of safe dosages
  • Recommend drugs that the doctors can prescribe to the patients based on the diagnosis provided by the doctor
  • Use PubMed to show the recent updates in research for that particular drugs and the keywords for the drugs
  • Generate a drug-disease model from the dataset

Built With

Share this project:

Updates