Mining AI and ML: How the Pharma Industry has leveraged Automation to Give Rise to Value-based Healthcare?

In recent years, technological, regulatory, and environmental changes along with supply chain imbalances have put a tremendous pressure on stakeholders to initiate automation in the life sciences industry globally. The application of AI and machine learning methods has produced demonstrable results in financial and manufacturing sectors and pharma leaders were quick to latch on to its usable benefits. But the industry was slow to adopt AI and ML due to the complexity of biology and availability of large and unstructured data sets alongside the costs of implementing such digitization practices in the short term.

Despite this, pharma was quick to respond to the pandemic and its impact. It caused companies to adjust to supply chain and clinical development disruptions that was previously unthinkable. Innovation and undisrupted drug supply alongside care delivery and financing also stimulated the growth in the adoption of AI.

Artificial Intelligence in the Life Sciences Market was valued at USD 1,255.3 million in 2020, and it is estimated to be worth USD 5,402.1 million by 2026, growing at a CAGR of 29.13% during 2021-2026. Additionally, the industry also witnessed a cost pressure with a greater need for productivity with disruptions causing new and innovative market players to enter the pharma space globally. AI has a particularly strong impact on the analysis of small system of interest specific datasets that can be utilized to improve drug development and personalized medicine. Clinical trial research can also benefit from AI as the extensive process can be vastly reduced with AI systems in place. One is by using advanced predictive analytics on a broad range of data to identify patients for clinical trials for target populations quickly and efficiently. Machine learning applications can also facilitate tasks such as calculating ideal sample sizes, facilitating patient recruitment and using medical records to minimize data errors.

AI systems can be used in drug discovery, medical diagnosis, biotechnology, clinical trials, precision and personalized medicine and patient monitoring alongside marketing and commercialization of drugs in the supply chain.

Drug Discovery:

Artificial intelligence has been making inroads in drug discovery for the past ten years. Biotech companies use the AI-first approach in more than 150 small-molecule drugs in discovery and more than 15 in clinical trials. AI can deliver value in small-molecule drug discovery in four ways: access to new biology and improved or novel chemistry, better success rates with quicker and cheaper discovery processes. AI can also address many challenges and constraints in traditional R&D. Each application brings additional insights to drug discovery teams and in some cases fundamentally redefine long standing workflows. These disruptive technologies are applicable to a variety of discovery contexts and biological targets.

Health Claims Fraud Detection:

AI is ideally suited to fraud detection for medical claims. Machine learning models can be used to automate claims assessment and routing based on existing fraud patterns. This process flags potentially fraudulent claims for further review but also added benefit of automatically identifying good transactions and streamlining their approval and payment. More advanced anomaly detection systems can be deployed to find new patterns and flag those for review leading to prompt investigation of new fraud types. AI systems can also provide clear reason codes for investigators so that they can quickly see the key factors that led the AI to indicate fraud which streamlines their investigation. With AI based fraud detection, fraudulent claims can be evaluated and flagged before they are paid thereby reducing costs for payers and keeps costs lower for patients and catching fraudsters in the act.

Data Quality Management System:

AI and ML has transformed data quality management by aiding data predictions and improving data quality by automating data entry through executing intelligent capture. This ensures that valuable data is captured and there are no gaps in the system. AI also helps in eliminating duplicating records in an organization’s database and helps keep precise gold keys in the database. AI-enabled intelligent systems can detect and remove duplicate keys in the data repository. AI-enabled system also removes defects in a CRM system. Data quality can also be enhanced through the implementation of machine-learning-based anomalies. AI also corrects and maintains data integrity. Third party organizations and governmental units can add significant value to the quality of a data management system and MDM platforms by presenting better and more complete data contributing to precise decision making. AI makes suggestions on what to fetch from a particular set of data and building connections in the data. AI helps clean data in one place and enables organizations of making informed decisions. There are leading data quality tools that use ML and are effective in data quality assessment and enhancement and have promising prospects to churn large data sets and enhance data quality.

Sales Forecasting and AI:

AI-enabled CRM systems are far superior to those that are not supported by this technology. AI synthesizes sales data in seconds and helps sales representatives to be more efficient and accurate when it comes to pre-calling, e-detailing and even guided selling. With AI, sales reps can make informed sales calls that lead to built-in competitive advantage and superior sales results. AI-powered systems can help sales reps make effective sales calls and create robust insights.

AI technologies can be used to leverage real world commercial and scientific information at early stages of drug development and research. These can be used to optimize clinical trial design and help predict and monitor risks. Simulations driven by AI can help researchers understand impact of key dynamics and plan for crucial scenarios slashing response times and improving the overall quality of clinical trials. AI (supervised and unsupervised learning algorithms) and predictive data analytics with machine learning allows researchers to process enormous amounts of data and information.

Predictive modelling of biological and environmental factors along with molecular and clinical data can help identify potential candidate molecules having the high probability of becoming safe and effective drugs for target populations. Robotic process automation also allows for more experiments to be performed within a given timeframe and minimizes the number of errors in clinical trial data capture. Other processes such as data entry into statistical databases, monitoring of protocol compliance and data quality checks can also be done by RPA with increased speed and precision. These processes free up valuable time that researchers and scientists can then use for high value work that need human creativity and attention.

In maintaining compliance challenges and privacy issues in complex drug regulatory environments, AI technologies can make data accessible for real-time data monitoring, investigation and audits across the R&D value chain. AI can help monitor and process vast amounts of data from multiple sources. Cognitive technologies such as NLP (natural language processing) and image processing can be used to automate time consuming manual work to improve data analyses and regulatory compliance and security.

These technologies help enable the quality and compliance functions that allow the monitoring of risks in real-time. Companies can then identify outliers and allocate resources efficiently. AI technologies can also improve safety surveillance, regulatory information management, compliance monitoring, structured content generation and continuous drug risk-benefit profile monitoring with reduced time and effort. These also help companies share accurate and timely information with regulatory bodies to market the drug faster. Machine learning and predictive analytics can also be used to identify right candidates for clinical research by analyzing a wide range of data that includes visits to physicians, social media interactions and population-specific information. Drug companies can then implement more targeted and efficient trials that lead to better adherence.

There are multiple drug therapies that leverage personalized medicine and can be devised by physicians with the help of supervised learning and AI capabilities. These help data scientists to draw conclusions from a limited set of diagnoses and individual health data. Pharma companies can also leverage insights generated from AI from patient interactions and dialogues to enable innovation and development of patient- centric services and products.

By analyzing longitudinal data, AI and ML can identify systemic issues in the pharmaceutical manufacturing process. These can highlight production bottlenecks and predict completion times for corrective actions and reduce the length of batch disposition cycle and investigate customer complaints. These technologies can also monitor in-line manufacturing processes to ensure safety and quality of medicines. Life sciences companies can improve their efficiency by applying AI to their supply chain management and logistics processes and align production with demand alongside AI-enabled sales and operations planning.

In the pre-clinical area, Novartis used AI in the initial stages of R&D to significantly increase drug compound identification, DNA interpretation and to scan the drug safety data. AI speeded up a high throughput screening. ML was used to manage genomic data and for the clinical trials, applied to examine trial data, predict adverse events and medical conditions, improve the predictiveness of diagnostic testing. Novartis has been instrumental in deploying QuantumBlack solutions that has reduced patient enrolment times by 10 to 15 per cent. The company has also entered a partnership with IBM to make use of IBM’s AI platform – IBM-Watson to improve clinical trial recruitment and make use of intelligent AI algorithms to predict medication efficacy.

Johnson & Johnson or J&J implemented AI since 2015 after collaborating with IBM’S Watson Health. The company is investing heavily in discovering precision medicine. It aims to provide personalized healthcare services to patients based on genetic profiles to improve patient outcomes and reduce healthcare costs. In collaboration with Winter light Labs, the company reaps the benefits of an AI platform that monitor’s neuropsychological details to treat Alzheimer’s. The speech-based smart AI platform can detect, understand and record hundreds of measurable variables from a user’s natural language. The automatic analysis of the disease is then identified.

By integrating IoT sensors into medical devices such as lenses and artificial joint replacement products, J&J plans to advance the clinical monitoring process of patients after a successful surgery.

Patients and end users are often eagerly engaged in finding relevant drug and illness related information through online research, word-of-mouth and social media interactions. However, a lot of the times, necessary information may not be easily available, and this can spurn customers from using a drug thereby affecting their treatment outcomes. This behavior is detrimental to the health and well being of pharma companies and AI can help companies bridge this divide by conducting personalized marketing initiatives based on online consumer behavior.

AI analytics can scour information on purchasing activity, amount of time spent on social media, nature of interactions on websites, apps and social media. It can provide predictions about ideal marketing messages, frequencies and media as well. Machine learning and natural language processing or NLP have been successfully leveraged to identify individual affinities and interactions on social media to deliver targeted marketing messages to the right audience. AI tools also leverage social media to increase transparency and positively impact patient behavior. NLP and computational linguistics alongside analysis of biometric data can provide insights into patient lives. These helps map sentiments behind reviews, ratings, comments and other online expressions.

AI and ML has varied uses in healthcare. For healthcare practitioners, ML and other cognitive disciplines can be used for medical diagnosis purposes. Using patient data and other information, AI can help doctors and medical providers with accurate diagnoses and treatment plans. AI also helps make healthcare more predictive and proactive by analyzing big data to develop improved preventive care recommendations for patients. With machine learning and AI, patient data can be gleaned for insights and help improve patient journeys. Machine learning systems offer an opportunity for hospitals to improve overall health outcomes. For many HCPs, AI can complete many manual tasks and help free up worker time. AI can also proactively identify a range of health conditions with faster speed and accuracy. AI-powered clinical surveillance can save lives and reduce conditions that have proven resistant to prevention.

While the benefits of AI are limitless, issues around data are at the heart of successfully promoting AI solutions in pharma. Healthcare is the least digitized sector and needs a systemic approach to develop common data standards and processes to maximize the value of existing data. Healthcare providers and AI companies need to put a robust data governance and ensure interoperability and standards for data formats, enhance data security and bring clarity to consent over data sharing. Another main challenge is the question of transparency. Since the complexity of processes involving artificial intelligence is quite a task, AI implementation is slow with pharma companies showing hesitancy to change.

As pharma companies move away from traditional, top-down product promotion and marketing healthcare model to a more flexible and interactive model where automation technologies are leveraged to enable patients to be more engaged in their health transformation, the move toward a personalized value-based healthcare model is more apparent. Rigid data silos have gradually been replaced with data that flows easily within functions and processes, integrating it at all stages of the healthcare and pharma value chain. Predictive analytics is generating business value and facilitating the shift to a personalized, attentive value-based healthcare unseen before.

The life sciences industry is facing a paradigm shift by transforming themselves to drive outcomes by transitioning to value-based care models. The patients and end users are significant beneficiaries of this better access to care. Driving this transition to value-based care is the use of personalized medicine catalyzed using artificial intelligence and data science helping reduce time-to-value at scale and in a compliant environment.