Merative
| File:Merative logo.svg | |
| Company type | Private |
|---|---|
| Predecessor | IBM Watson Health |
| Founded | June 30, 2022 |
| Headquarters | Ann Arbor, Michigan |
Key people |
|
| Owner | Francisco Partners |
Number of employees | 1,000–5,000 |
| Website | merative |
Merative L.P., previously known as IBM Watson Health, is a U.S.-based healthcare technology enterprise delivering solutions and services that aid medical research, clinical research, real world evidence, and the broader Healthcare industry by utilizing artificial intelligence, advanced Analytics, cloud infrastructure, and sophisticated IT systems. Owned by San Francisco's private equity firm Francisco Partners, Merative emerged in 2022 following IBM’s divestiture of its Watson Health segment. The company functions independently with its main headquarters in Ann Arbor, supported by innovation hubs located in Ireland, Hyderabad, Bengaluru, and Chennai.
History
Originally part of the Thomson Corporation, Thomson Healthcare became the healthcare division of Thomson Reuters after their 2008 merger. On April 23, 2012, Thomson Reuters agreed to divest this healthcare unit to Veritas Capital for $1.25 billion USD. This transaction concluded on June 6, 2012, resulting in the establishment of Truven Health Analytics as an autonomous healthcare data company.
On February 18, 2016, IBM acquired Truven Health Analytics for $2.6 billion and merged it into its Watson Health division. Truven was known for delivering extensive healthcare data and analytical services. The brand name "Truven" reflects a combination of "trusted" and "proven."
In January 2022, IBM declared the sale of segments of Watson Health, including Truven, to Francisco Partners for over $1 billion. The acquisition was finalized on June 30, 2022, when Francisco Partners launched Merative as a specialized healthcare data organization.
Advancements
Watson’s capabilities in natural language processing, hypothesis formulation, and evidence-based learning are being developed to improve clinical decision support systems and broaden the scope of artificial intelligence in healthcare tools available to clinicians. When a doctor inputs symptoms and related clinical data, Watson extracts critical details, analyzes patient histories, reviews multiple data sources to generate and assess hypotheses, then delivers tailored treatment recommendations with confidence scores. Its information sources encompass treatment protocols, electronic medical records, provider documentation, scientific research, clinical trials, academic journals, and patient data. While promoted as a "diagnosis and treatment advisor," Watson's role is limited to assisting with treatment options for patients who already have a diagnosis.
In February 2011, IBM collaborated with Nuance Communications to create a commercial product that harnessed Watson’s clinical decision support functions. Medical experts from Columbia University and the University of Maryland helped identify key clinical challenges and optimal ways for Watson to support healthcare providers.
By September 2011, IBM partnered with WellPoint (now known as Anthem) to utilize Watson’s analytic abilities in recommending treatment plans. By February 2013, Watson was applied commercially for utilization management decisions regarding lung cancer care at the Memorial Sloan–Kettering Cancer Center.
In October 2012, IBM announced an alliance with the Cleveland Clinic to implement Watson at the Lerner College of Medicine at Case Western Reserve University, aiming to improve diagnostic accuracy and speed through advanced data processing.
In 2013, IBM and the MD Anderson Cancer Center launched a pilot initiative focused on cancer treatment innovation; however, after investing $62 million without achieving targeted outcomes, the program was terminated.
On February 8, 2013, oncologists at the Maine Center for Cancer Medicine and Westmed Medical Group in New York began evaluating Watson’s recommendations for lung cancer therapies.
On July 29, 2016, IBM introduced Watson for Oncology in partnership with Manipal Hospitals in India to deliver personalized cancer care insights grounded in evidence-based medicine. Manipal Hospitals became the second institution globally to implement this technology and the first to offer it online as an expert second opinion service. This agreement concluded in December 2018.
On January 7, 2017, IBM signed a contract with Fukoku Mutual Life Insurance to deploy IBM Watson Explorer AI for analyzing compensation claims. This deployment resulted in a reduction of workforce by 34 positions, productivity gains of approximately 30%, and operating cost savings amounting to ¥140 million.
IBM Watson is anticipated to consolidate the expertise equivalent to that of 1,000 oncology specialists, representing a transformative advancement within healthcare; nonetheless, its applications in oncology are still emerging.
Entrepreneurial ventures in healthcare have adopted seven distinct business model archetypes based on IBM Watson technology to deliver solutions targeting either patients or providers through diverse value creation and capture strategies.
By 2019, evaluations characterized the Watson Health initiative as an example where expectations exceeded actual outcomes. A 2021 commentary from the Association for Computing Machinery (ACM) underscored challenges related to portfolio management within Watson Health due to multiple acquisitions since 2015 and an emphasis on branding rather than addressing client requirements.
On January 21, 2022, IBM revealed plans to sell its Watson Health assets to private equity firm Francisco Partners.
Industry considerations and challenges
Consolidations among major healthcare corporations improve access to medical data, enabling wider adoption of artificial intelligence technologies. A central industry priority is incorporating AI into clinical decision support tools, which enhance operational efficiency amid escalating data volumes. Numerous organizations are harnessing big data analytics to drive innovations in healthcare.
IBM's Watson Oncology is being developed in partnership with Memorial Sloan Kettering Cancer Center and Cleveland Clinic. IBM also collaborates with CVS Health to apply AI in managing chronic illnesses and works alongside Johnson & Johnson to mine scientific publications for drug discovery insights. In 2017, IBM and Rensselaer Polytechnic Institute initiated the Health Empowerment by Analytics, Learning and Semantics (HEALS) project, aimed at advancing AI applications in healthcare.
Additional leading companies have contributed AI algorithms tailored for medical use cases.
Microsoft
Microsoft's Hanover initiative, conducted with Oregon Health & Science University's Knight Cancer Institute, employs AI to study medical literature and forecast effective cancer therapies. Other projects include analyzing tumor progression through medical imaging and engineering programmable cells for therapeutic purposes.
Google's DeepMind collaborates with the UK National Health Service (NHS) to identify health risks by analyzing data from mobile applications. Another partnership with the NHS focuses on developing computer vision algorithms for cancer detection by examining medical imaging data.
Intel
Intel Capital has invested in Lumiata, a startup utilizing AI to pinpoint patients at risk and customize care plans accordingly.
In healthcare, artificial intelligence applies sophisticated algorithms designed to mimic human reasoning when interpreting clinical data. AI systems differentiate themselves by processing information and producing accurate results via machine learning models that detect patterns and apply logical inference. These models undergo rigorous validation to reduce errors and contrast with human cognition by strictly following explicit instructions without interpretation, functioning as "black boxes" that provide precise predictions without revealing underlying rationale.
The foremost business objective of AI in medicine is to analyze correlations between preventive or therapeutic interventions and patient outcomes. AI's applications encompass diagnosis, development of treatment protocols, drug discovery, personalized therapies, and continuous patient monitoring. Prestigious institutions such as The Mayo Clinic, Memorial Sloan Kettering Cancer Center, and the National Health Service have integrated AI-driven algorithms into their clinical workflows. Prominent technology companies including IBM and Google, along with innovative startups like Welltok and Ayasdi, play significant roles in advancing healthcare AI solutions.
Healthcare facilities increasingly implement AI-powered predictive analytics to streamline operations, reduce expenses, enhance patient experiences, and tackle staffing challenges. These tools assist administrators in improving resource utilization, minimizing patient wait times, shortening hospital stays, and optimizing workforce deployment.
The following medical specialties represent key areas of focus for artificial intelligence research:
Radiology
AI-powered analysis of medical images assists healthcare professionals in identifying subtle abnormalities that may otherwise go unnoticed. Research from Stanford introduced an algorithm that detected pneumonia with a superior average F1 score—a combined measure of precision and recall—compared to radiologists involved in the study. The annual conference of the Radiological Society of North America now regularly includes presentations focused on AI applications in imaging. Despite measurable improvements in certain performance indicators, some radiologists remain cautious, viewing AI as a potential challenge due to its isolated successes relative to human expertise.
Imaging
Recent AI developments are utilized to assess outcomes in maxillofacial surgery and cleft palate treatment by evaluating elements such as facial esthetics and perceived age. A 2018 study published in the Annals of Oncology revealed that a deep convolutional neural network identified skin cancer with 95% accuracy, surpassing dermatologists who achieved an average accuracy of 86.6% when diagnosing through images.
Disease Diagnosis
Artificial intelligence has been successfully employed to detect prevalent diseases like Diabetes and Cardiovascular Disease (CVD), which rank among the leading causes of death worldwide. Due to their significant mortality rates, substantial research efforts aim to improve diagnostic precision through AI technologies.
Jiang et al. (2017) reviewed multiple AI methodologies applied in healthcare, including Support vector machines, neural networks, and decision trees, all designed to enhance classification performance. Notably, two primary approaches for disease identification are "Artificial Neural Networks (ANN) and Bayesian Networks (BN)."
An analysis of studies conducted between 2008 and 2017 found that machine learning techniques such as ANN and BN facilitate early disease classification. Among these, ANN exhibited the highest accuracy rates for diagnosing diabetes and CVD, with mean accuracies of 87.29% and 89.38%, respectively.
Telehealth
The growth of Telemedicine has created new opportunities for integrating AI technologies. AI-driven patient monitoring systems can notify clinicians about potential disease progression by analyzing continuous data streams.
Wearable technology allows for ongoing health surveillance, capturing subtle physiological changes that may escape human detection, thereby supporting more proactive care and effective remote health management.
Electronic health records
Electronic health records (EHRs) play a crucial role in healthcare digital transformation and information sharing. However, the extensive documentation required can contribute to cognitive strain and professional burnout. To mitigate these issues, EHR platforms are increasingly incorporating automation for clinical documentation alongside natural language processing (NLP) tools. Research from the Centerstone Research Institute demonstrated that predictive models based on EHR data achieved 70–72% accuracy in forecasting personalized treatment outcomes, suggesting AI’s potential in reliably identifying disease etiologies from EHR information.
Drug Interactions
Progress in Natural Language Processing has led to the development of algorithms capable of detecting drug-drug interactions within medical literature. Such interactions represent significant risks, particularly for patients on complex medication regimens. Machine learning techniques extract relevant interaction details from texts, supported by initiatives like the 2013 DDIExtraction Challenge organized by Carlos III University, which provided a standardized dataset for evaluating algorithm effectiveness. This dataset continues to serve as a foundational benchmark for assessing extraction accuracy.
Further algorithms analyze user-generated sources including electronic health records and adverse event reports to uncover drug interaction patterns. Regulatory agencies such as the FDA Adverse Event Reporting System (FAERS) and WHO’s VigiBase collect data on medication-related reactions; deep learning models process this information to identify interaction trends, thereby enhancing pharmacovigilance efforts and improving patient safety.