Transforming Patient Health: The Power of Data Science in Pharmaceuticals
By leveraging Data Science, AI, and other digital technologies, the healthcare industry could build complementary health solutions that are personalized to the specific needs of patients. Here is how and why.
The world population grows by more than 80 million per year, according to a 2017 report by the United Nations. By 2050, there will be 10 billion people on this planet and people over the age of 60 years and above are expected to double. It’s clear that a growing and aging world population needs better and more sustainable solutions for health and nutrition. Data Science has the power to fundamentally transform patient health and the healthcare industry needs to leverage digital technologies for better solutions.
Artificial Intelligence and Machine Learning help to leverage data so patients can be diagnosed earlier and to gain a thorough and deep understanding of diseases. This is key to tailor treatments to the individual needs of a patient and to identify those patients who will benefit the most. Moreover, data-driven digital solutions help to bring new medicines to patients faster than ever before.
Integrated patient care starts with identifying relevant information. Diagnosing a disease early on can have a significant impact on the outcome. Digital solutions could help healthcare professionals to make an appropriate diagnosis as early as possible.
For example, a software could use deep learning methodology to support radiologists in identifying signs of CTEPH (chronic thromboembolic pulmonary hypertension), a rare form of pulmonary hypertension, in CTPA (computed tomography pulmonary angiogram) scans. The software processes image findings of cardiovascular, lung perfusion and pulmonary vessel analyses in combination with the patient’s history of pulmonary embolism.
Also, an AI algorithm could help doctors identify patients with a high risk of cardiovascular diseases (such as heart failure and recurrent stroke) earlier and with more precision than ever. Patients could be differentiated based on complex individual profiles made up of a unique combination of characteristics (demographics, and clinical risk factors such as diabetes and genotypes). This requires the application of outcome data from clinical studies, genomic and imaging data for this stratification.
Individualized patient treatment is a core ambition that can be delivered through data. Another line of work seeks to tailor treatments to the underlying cause of an individual patient’s disease. One such example is the development of an AI algorithm that is intended to identify patients whose cancer is likely the result of an NTRK gene fusion in their tumor cells – often resulting in an altered TRK fusion protein, leading to cancer growth. While overall rare, TRK fusion cancer affects both children and adults and occurs in varying frequencies across various tumor types, which makes testing for this alteration so important. The AI algorithm may help physicians to identify patients who are likely to have TRK fusion cancer, based on their tumor pathology. These results are then verified by the specific validated diagnostic methods that are already used.
