Can You Find Your Patients? Using AI for Faster Patient Identification and Treatment

Artificial Intelligence is a valuable tool to aid pharma in both finding where their patients are (which physician) and aiding the physician in faster diagnosis if the condition is one that is more difficult to diagnose.
This is especially valuable in rarer conditions where money is spent educating a significant number of physicians who will never see a patient in their lifetime, but equally so in all conditions if we are spending valuable resources targeting physicians who do not have that patient population. In addition, AI is also used when it is difficult to identify patients due to generalizes symptoms and to diagnose a patient rapidly (e.g. in Multiple Sclerosis for whom patients are often not diagnosed by their primary care practitioner for around 10 years before they are sent to a neurologist and by then, much irreversible damage has taken place leaving the drugs less effective).

Now, by using AI, in all these cases, we can identify all of these patients within the data sets (EHR and claims data) within minutes (after the initial time spent data wrangling and creating the algorithms). Then every time a new patient enters the healthcare system, that patient can be immediately identified using the algorithms.

Finding specific patients to precision-target physicians

In rare disease patients, Eularis have been able to identify specific rare conditions using face recognition from photos uploaded online with far greater than expected results.

In other conditions, such as specific cancers, we can utilize AI to identify which patients have the condition from their patient records (for example, if the condition is 3L then we can identify both which patients are currently second line and have a high prediction of progressing to 3L, which allows us to know which physicians the reps should be seeing to prepare them for the patient).

In any condition, we can use AI on EHR and claims data to identify the patients (not personally as PII is taken out ) but we can identify which physician will be seeing them, and aid the reps in physician targeting as we can precisely identify which physicians are the ones who will be seeing patients needing your drug, who will be needing to write scripts for your condition very soon, and which are the most appropriate for discussions on the suitability of the product for their patients.

Finding patients to speed up diagnosis
In the past, delays in diagnosis due to the rarity of the conditions, often leading to exacerbated severity, meant many patients were never diagnosed, or the condition was so late stage by the time they were, it did not help them. The average time taken to diagnose a rare disease without technology is 7.6 years (according to Shire Pharmaceuticals) after countless tests and physician visits. This causes a lot of unnecessary costs to the healthcare system and much suffering for the patient.

Now, using AI, we can identify all of these patients within the data sets (EHR and claims data) within minutes after the initial time spent data wrangling and creating the algorithms. Then every time a new patient enters the healthcare system, that patient is immediately identified using the algorithms.

Finding patients to speed up clinical trials
Patient recruitment and retention is frequently the most labor intensive and difficult part of the clinical trial process and is often a barrier to the successful completion of a study. Industry data indicates that patient recruitment can represent more than 30 % of total study costs and sponsors bear a tremendous cost when ambitious recruitment goals, often misguided, fall short. In recent years, a number of technology innovations, such as eConsent and patient portals, have been adopted by pharmaceutical companies and CROs to aid this process. However, a further advancement is to use Artificial Intelligence to identify appropriate patients for clinical trials. Technology solutions, such as using patient registries and online forums, are also playing a part in the transformation of this area by being used to directly recruit patients and improve patient retention by providing an additional communication channel to standard doctors’ appointments.

Using AI and combining this with clinical and laboratory data from Electronic Medical Records (EMR) can also identify suitable investigators and patients. Patient profiling is used to create a profile of the typical patient for each trial, which helps to target the correct patients.

Besides cutting costs, improving trial quality, and reducing trial times by almost half, AI can be used to find biomarkers and gene signatures that cause diseases, identifying eligible clinical trial patients in minutes by reading volumes of text in seconds.

Using EMR and other data, Eularis utilize Artificial Intelligence to automate:
    •    the cumbersome process of identifying patients who meet the criteria for a clinical trial,
    •    analysis of protocols to assess trial feasibility
    •    identification of optimal trial sites.

This approach can rapidly identify how many patients match a clinical trial’s inclusion criteria, where the patients are located and how they will be recruited.

The Eularis approach of applying AI to this facilitates access to anonymized patient data by enabling the massive amount of personal health data being created daily to be anonymized, shared, and combined with a dynamic and constantly growing aggregated view of clinical, research, and social health data. This allows better access to patient populations and the increased information available enables site and patient recruitment to be optimized, with improved matching of patients for clinical trial participation