Topics In Demand
Notification
New

No notification found.

7 use cases of AI in Healthcare that delivers immediate value
7 use cases of AI in Healthcare that delivers immediate value

373

0

#AI in Business

1. AI-Search for Medical Records (unlock the gold mine)

Typical usage of Electronic Medical Record (EMR) systems implemented in India is to

(a) capture the medical records such as medical history, investigation reports, and treatments (b) query the system for analysis.

The medical records captured are semi-structured, non-standardized (e.g., HL7), non-coded (ICD-10) and consists of free-text (as doctor notes, discharge summary) and pdf reports. Sometimes, the data resides in silos (e.g., inpatient or outpatient data). These silos cause challenges in querying the systems for rich information that is required to generate insights. Along with this, the EMR providers offer very standard interfaces/reports for querying. Hence, the rich medical data that can provide deep insights cannot be unlocked.

For instance, a doctor may like to query EMR to gather insights from previous treatments for some rare or complex medical cases. A doctor may further like to research on these topics for clinical journals.

AI-Search for medical records leverages the existing EMR and AI technology, such as clinical NLP, to extract "entities of interest" from records. Along with this, the system keeps learning relationships with the associated text and builds a knowledge graph.

For example, consider a case where a doctor may want to search for "kidney failure multiple myeloma." Some doctors may write this as "renal failure, MM" or "auto-immune, kidney-failure" as these are all associated texts. It would not be straightforward to find this through a regular search or query engine but can be easily achieved with AI Search.

The next phase of AI-powered Search would be for the doctors to ask queries in natural language. For example, for the search mentioned above, doctors may query, "Find me all the kidney failure cases for people without diabetes."

Benefits: Doctor's engagement, Improved health outcomes, Improved clinical capability

2. Treatment (Hospitalisation) Pricing Prediction

The treatment cost for certain complex medical conditions can vary a lot depending upon the co-morbidity and other factors. AI can provide an indicative length-of-stay and estimated cost for the patients. A data scientist may use historical data for the treatment cost, and all the associated medical conditions to develop such a prediction model. The model can further offer a fixed-price package for treatments as opposed to the current room-rent based model.

Benefits: A new pricing model, revenue forecasting

3. Predicting Patient Flow: Bed Utilisation, Emergency and OPD traffic

Capacity planning to improve operational efficiency and prevent any revenue leakage is a crucial metric for hospitals. Hospital patient-flow consists of admitted, scheduled, and unscheduled patients. Basis the historical data and alternate data such as weather, environment quality, local events a predictive model can be developed that can model the flow.

Benefits: Revenues and asset utilisation, improve efficiency

4. Human Detection in Restricted Area or Infectious Zones

There may be hazard-zones in a hospital that prohibit humans from entering the area. With real-time camera feeds as inputs, AI-models can automatically detect any human presence. These models can be useful in highly secure environments or infectious wards to raise an appropriate alarm automatically.

Benefits: Safety, Compliance, Infection control

5. Hospital Incident Prediction

Hospital incidents such as infections, sepsis, transfusion reactions pose significant operational, clinical, and reputation risks to both patients and hospitals. For controlling such events, hospitals implement robust Standard Operating Procedures (SOPs) & checklists. At times due to the volume of at-risk patients or other variables, the chances of occurrence of hospital incidents increase.

By analyzing the real-time patient profiles, AI models can predict such anomalous situations. Accordingly, warnings can be issued to hospital management, allowing them enough time to take preventive measures.

Benefits: Infection control, Incident control

6. Patient & Brand Sentiment Analysis

Patient satisfaction is a key performance indicator not only for the hospital but also for future patients. An AI model trained on patient feedbacks (forms, audio or paper-based) and open consumer reviews on the internet for hospitals or doctors associated with the hospitals can derive an overall brand sentiment for a hospital. Such a brand sentiment may make it easier for future patients to visit the hospital of their choice. 

Benefits: Infection control, Incident control

7. Intelligent Marketing

Hospitals (HIS) have large data-sets of patients (customers). The usual way of slicing the data is basis the disease type, department, age, gender, consulting doctor, treatment availed etc.

Using "clustering" techniques, we can leverage AI to "Tell me what pattern exists in my data". This can identify common characteristics or attributes to explore costs, treatments, or outcomes, which the hospitals can use for marketing, designing newer programs etc

Benefits: New Business


That the contents of third-party articles/blogs published here on the website, and the interpretation of all information in the article/blogs such as data, maps, numbers, opinions etc. displayed in the article/blogs and views or the opinions expressed within the content are solely of the author's; and do not reflect the opinions and beliefs of NASSCOM or its affiliates in any manner. NASSCOM does not take any liability w.r.t. content in any manner and will not be liable in any manner whatsoever for any kind of liability arising out of any act, error or omission. The contents of third-party article/blogs published, are provided solely as convenience; and the presence of these articles/blogs should not, under any circumstances, be considered as an endorsement of the contents by NASSCOM in any manner; and if you chose to access these articles/blogs , you do so at your own risk.


images
Vikas Gupta
Co-Founder

© Copyright nasscom. All Rights Reserved.