Artificial Intelligence – Critical Prerequisite of Clinical Trials

A vast majority of researchers and specialized tech startup companies have started investing in developing Big Data and Artificial Intelligence (AI) tools to serve the pharmaceutical and medical devices companies, which is believed to transform the clinical trial process. But what exactly is the main outcome of AI?

The buzz around AI in clinical trials is due to its potential as the lynchpin for dramatically improving the probability of success and reducing the timelines.

How does AI really increase Clinical Trial Success Rates?

The answer lies in predictive analysis from available historical data. The main idea of the AI revolution is to bring about “efficient and faster decision making,” provide precision in clinical trials and to bring an effective product from the lab-to-market. The average timeline for a drug molecule to be released from lab-to-market is 9 years with a median development cost of $2 billion. The objective of AI implementation is to eliminate unnecessary repeated clinical evaluations, save costs & time and thereby ensure successful clinical trials.

AI based Clinical Trial transformation process can be divided into three main components:

Innovative Study Design

The first and foremost prerequisite for AI implementation is data mining.

• Cross-comparison analysis of historical similar molecule protocol designs
• Centralized Monitoring of Trial Risks
• Measuring Drug Responses, Trial Outcomes, and Site Performance Prediction

Patient Recruitment and Retention Metrics

The optimum utilization of patient data will ensure effective patient recruitment and lower dropout rates. Several AI startups such as “DEEP6” and “antidote” have invested in developing patient trial matching software.

• Tapping EMR/EHR to match the right patient trials
• Patient eligibility Good-fit analysis for trials and predicting precise responsiveness

Smart Data-Driven Analysis

Artificial intelligence is now being used for predicting cancer treatment type based on the combination of genes in the clinical trials right in the planning stage.

• Predictive result analysis based on the drug molecule behavioral pattern
• Dynamic Real-time access to data and sharing as and when there are updates
• Effective Go/No-Go decision for every stage of the clinical trial process

Leverage AI in Clinical Trials with Salesforce Einstein Platform Enablement

The Salesforce Einstein platform can act as a smart assistant which can be integrated into a CTMS. The Einstein Voice and Einstein predictive builder can help leverage artificial Intelligence in clinical trials in the following ways:

• Discover insights: The Einstein Discovery feature enables CROs and Sponsors to analyze every data point, pulling together data from external sources, to create an effective study design in the planning stage of a clinical trial.
• Predict outcomes: The Einstein Prediction Builder feature can be used to build a customized predictive score model to analyze required patient data tapped from EHR / EMR. Based on the analysis score, the closest trial outcomes can be predicted to match the patients suitable for the clinical study.
• Recommend: The platform also enables users to analyze site performance metrics, which will help in recommending appropriate sites for selection based on responses.
• Automate routine tasks: All monitoring activities for scheduling monitoring visits, audit trails, approval of documents, reporting and updates based on recommendations can be automated at any stage with Cloudbyz Centralized RBM solutions. All the activities and reminders can be set up hands-free using Einstein Voice feature.

The Final Word

Investment in AI and Big Data will certainly improve the clinical trial process. However, initial investments on tools and technologies need to be taken into consideration. Further, market experience with AI tools will bring down the costs and time-to-market of the drug. How AI can transform the drug discovery process, and make the above statement true – Only time will tell.

References

https://www.himss.org/library/ehr

https://www.himss.org/electronic-medical-records

https://www.ijclinicaltrials.com/index.php/ijct/article/download/125/94

https://www.jpmorgan.com/commercial-banking/insights/ai-transform-clinical-trials

https://healthcareweekly.com/artificial-intelligence-clinical-trials/

http://www.iagrp.com/views/3_types_artificial_intelligence_clinical_research/

http://www.clinicalinformaticsnews.com/2017/09/29/the-intelligent-trial-ai-comes-to-clinical-trials.aspx

https://www.salesforce.com/in/products/einstein/overview/

https://www.salesforce.com/video/193342/

https://www.salesforce.com/products/einstein-analytics/products/einstein-discovery/

https://www.salesforce.com/products/einstein-analytics/products/einstein-discovery/#prediction-builder

https://www.salesforce.com/products/einstein/einstein-voice/