Home  >  Special Features
Eppen_MCX50_Dec2022
you can get e-magazine links on WhatsApp. Click here
Special Features
+ Font Resize -

Role of artificial intelligence in clinical trials

Dr. Upasana Shukla
Tuesday, June 7, 2022, 08:00 Hrs  [IST]

Artificial intelligence (AI), the term coined by computer and cognitive scientist John McCarthy in 1955, is the combination of science and engineering to mimic human intelligence including learning, reasoning, and perception. The use of AI in the drug development process has proven to be one of the most significant technological achievements of the 21st century. Given the ability of AI to generate accurate data collection and efficient data management, it is poised to play a major role in pharmaceutical research. With the perfect blend of revolutionary advances in computational technology and previous constraints of the collection of large volumes of data, AI techniques are appealing to the pharmaceutical industry. As AI has the inherent capability to predict the output precisely, along with its ability to process big data, it can enhance and elevate the existing pharmaceutical processes to new heights.

Clinical trials (CTs) are protocols developed to diagnose, treat, and prevent various clinical diseases and conditions. The majority of CTs are involved in the investigation of the safety and efficacy of a drug molecule. To launch a single drug in the market takes more than a decade and billions of dollars expenditure on research and development. Ironically the success rate of these lengthy clinical trials is less than 10%. The most recent field of drug development that has acknowledged and opened its gate for positive disruption from AI is clinical trials. The key factors for the success of any clinical trial are the clinical trial designs, patient cohort selection and recruitment, study site, investigator selection, patient monitoring, protocol adherence, and the outsourcing of required skills and talents. AI can automatically allow efficient patient selection through a wider range of data sources including Electronic and Medical health records. It helps biopharma companies to identify qualified investigators as well in patient management by automated data capture and sharing data across systems. As more and more data are electronically available, it is natural that AI will play an increasingly important role in mining the data efficiently. AI has the prospect to increase the likelihood of success in drug development by bringing significant improvements in multiple areas of R&D such as novel target identification, drug candidate selection, biometrics data analysis from wearable devices, and the prediction of the drug effects in patients with diseases.

The results obtained using AI can be applied in the creation of structured, standardized, and digital data elements from a range of inputs and sources. The transformation tool can leverage existing systems to seamlessly integrate the data flow resulting in providing a single, collaborative touch point for all interactions during a clinical trial. Using AI-enabled technologies, the researchers can investigate and generate insight from past and current trials, analyze enormous data, and inform the required adjustment in future trial designs. Beyond clinical trials, AI has great potential in the diagnosis of diseases and treatment applications as well. Using this smart machine-led predictive technology, the existing data based on standards can be reused and the requisite to start the trials from scratch can be reduced.

The growing body of academic research has demonstrated numerous applications of AI which can be beneficial in conducting clinical trials such as interpreting chest radiographs, detecting cancer in mammograms, identifying brain tumors on magnetic resonance imaging (MRI), and predicting the development of Alzheimer’s disease (AD) using positron emission tomography (PET) scans. Many orphan diseases such as neurodegenerative diseases which do not have effective treatments can be benefitted from the usage of AI in the field of genetics. For example, in the case of AD, genetic factors contribute to the major etiologic role. Currently, AI is being used for the diagnosis, gene expression, gene-gene interaction, and analysis of genetic variation of AD. The continuous improvement in the strategies of AI might prove as the steppingstone toward decoding this complex interaction of genetic variations. AI plays an important role in personalized medicines as well which involves disease treatments and prevention based on individual variability in genes, environment, and lifestyle of each person. The range of medical care covered by AI is from diagnosis, preventive medicine, and palliative medicine to drug design and development.

In the field of R&D, AI has the prospects of tremendous growth and potential, as with most technological developments, this also presents both challenges and hope. Retrospective studies have been majorly used to test and train the algorithms whereas only through the prospective studies the true utility of AI systems will be understandable. The major challenge is the generalization of AI systems to new study populations and settings due to technical differences as well as clinical and administrative practices. One of the side effects of the implementation of AI in clinical trials is bias within the data which can negatively impact the decision making. The unavailability of most healthcare data for machine learning is also one of the concerns in implementing AI systems. On the other hand, the time and cost saved by implementing AI in clinical trials cannot be ignored. Round-the-clock availability and zero human errors are some of the perks of AI. Consequently, increasing data volume along with various facets associated with data makes AI a natural choice to efficiently manage and mine the data.

Though AI methods represent major technological advances, if the organizations misread or misinterpret the confounding factors then it can lead to different conclusions which can be proven hazardous to the industry. Hence, it is vitally important to have proper knowledge, use reliable algorithms and data, and perfectly include clinical issues behind all the endpoints and data collection. Moreover, whatever the advances of AI in the healthcare industry, human supervision is still required. It is too early to expect the machine to understand the social variables, to have the human behavioral insights which are important to identify the pattern or avoid unexpected medical issues, and to be free from all security risks.

Given the pros and cons, still, for the next few years, clinical trials will be the benchmarks for understanding the safety and efficacy of clinical drug molecules and the use of AI in these trials will be indispensable. Machine learning and AI have tremendous potential for drug design improving both endpoints and adherence. It will enable the design and execution of clinical trials faster, more dependable, and patient-centric. Though predicting the future of AI, machine learning and other implications in clinical trials are very subjective and tricky as AI is still in its infancy. AI-enabled predictive analysis in the health sector is going to provide new opportunities. It will open the door for healthcare management of patients remotely without the need to visit the hospital, and even in hospitals, it will be able to reduce the waiting period. Artificial intelligence holds greater promises, not only in transforming clinical research, but also in reducing the cost associated with disease management, successful aging, and the discovery and development of new medical innovations.

Conclusion
A proper understanding of both R&D and advanced AI techniques can provide large benefits to the drug development industry and in the management of patients’ care. With proper execution and visualization of AI, it can offer user-friendly platforms to maximize efficiency and encourage the use of revolutionary techniques in clinical trials. With the inclusion of AI, big medical institutions can shift to a smaller AI-enabled standard care which will require minimal space from large teams working across dozens of data systems. The drug development process can be accelerated by AI-generated data collection and management which will help companies to provide new treatments to market more quickly, by reducing the time and effort required for clinical trials. It can be a gamechanger in personalized medicine, and the potential gains to be made in the creation of novel drugs and treatments. AI puts the power in the hands of life sciences and health care organizations which can provide patients faster access to safe medicines and even save their lives.

(The author is medical reviewer, Healthminds Consulting Pvt Ltd)

 

Post Your commentsPOST YOUR COMMENT
Comments
* Name :     
* Email :    
  Website :  
   
     
 
truking"
 
 
Copyright © 2016 Saffron Media Pvt. Ltd |