Docviser.

Docviser is an oncology smart system prescription and patients management tool that enables doctors to collaborate very efficiently without a necessarily physical presence using virtual medical committees (tumor boards) and a collaborative treatment protocol database.

The entire in-clinic prescription flow is handled between doctors > pharmacists > nurses that enables them to set the right status for every treatment with real-time visibility.

Create a collaborative tool for doctors to work on a global treatment protocol database.

Create a prescription system that enables doctors to prescribe efficient treatments with exact dosages calculations, choosing from a shared / global database of protocols.

Enable doctors, pharmacists and nurses to collaborate on each prescription and administer drugs efficiently without confusions and too many interactions.

Enable doctors to collaborate across departments (oncology, surgery, cardiology, anatomo-pathology etc.) without physical presence and make the best decisions for their patients.

Create a Machine Learning system that will collect relevant information about patients and prescriptions, patients' statuses and other genetic data markers to create a treatment forecast for recommended protocols.

Create a tool where doctors can create and manage patients' consultations, either online or on-site, and also automatically schedule Cycle Due consultations based on the prescribed treatments.

Implement a Chat GPT-like AI assistant that can answer questions about the patient’s uploaded files as well as check the interaction between the prescribed treatment and the co-medication that the patient is already taking.

Implemented a perfectly designed tool for doctors to create a shared database of treatment protocols that allows them to enter all the relevant data like drugs with active substance, cycles, days spread, dosages, custom calculations and localizations, in order to be used while prescribing a treatment. Uses ICD-11 or ICD-10 database, custom cancer localizations, bibliography, and an internal chat for doctors to collaborate and dispute certain protocols that are not entirely clear.

Implement a smart and efficient treatment prescription system that uses the treatment protocols database, scientific proven calculation methods for doses and other relevant blood tests samples to recommend the best suitable protocol.

We implemented the in-clinic flow that enables doctors to prescribe treatments, validate them and send them to pharmacists when they are ready. Pharmacists will then receive them in real-time either by web app notifications or via mobile app push notifications systems and will be able to change the status and send it further to nurses for administration. The entire cycle is visible by everyone in real-time and uses a chatroom to efficiently collaborate on every prescription.

We implemented a very performant system for Virtual Medical Committees or Tumor Boards. The system involves sharing relevant case study/patient data in a secure and structured environment for doctors in different specialisations to understand, discuss and make the best decisions for a specific patient. Doctors have the ability to schedule a secured and trusted online meeting with other doctors in a Google Meet style conference that supports up to 30-40 people with audio and video at the same time with adaptive streaming based on their bandwidth. They can use the Tumor Board chatroom, invite guests, share important files, and have a full-blown conference all in a secured one place, while navigating through the patient’s files and treatment history.

Patients can be scheduled for consultations, either online or on-site, and the system will automatically send the patient a notification with the scheduled time and date. The system will also automatically schedule the next Cycle Due on chemotherapy, based on the prescribed treatments. Patients will have access to their files, treatments and shared information in real time, and will be able to see the status of their treatments and the next scheduled consultation.

We designed and implemented a Google Calendar-like scheduler, where all users can schedule and manage patient consultations, either online or on-site. Furthermore, Pharmacists and Nurse users can see the upcoming prescriptions in a calendar view, to more efficiently manage the drug preparation and administration process.

The content of all the patient’s files as well as all the treatment history from Docviser was used in order to create a Vector Database for each patient, and using the GPT-4-turbo, GPT-4o, MedPub and other open-source models, we unlocked the possibility of interrogating the AI assistant about the patient’s files, auto-complete information in the patient’s medical files & epicrisis and check the interaction between the patient’s prescribed medication. Annonymized data was used to train the AI assistant, and the system is GDPR & HIPAA compliant.