Isabel produces a list of possible diagnoses, ranked in order of degree of match between the disease presentations and clinical features, together with age, gender and region, entered by the user. Clinicians can review the Isabel list and access associated evidence-based content to work out which diseases are most probable for their patient. Isabel is there to support and assist the clinician in determining the differential and working diagnosis. Isabel is about possibilities while clinical probability is determined by the clinician.
The clinical features can be entered in any order. It’s best to ensure you enter each clinical feature on a separate line. Clinical features can be entered in free text or selected from the predictive text drop down list presented.
In addition to signs and symptoms, other clinical features such as abnormal test results, family history, vital signs, social history and comorbidities can be entered into Isabel. When entering lab values convert numerical data into the textual meaning for example, Hemoglobin 4.2mmol/l into ‘low hemoglobin’.
Over the past 15 years that Isabel has been available and used by clinicians around the world we have not received any reports of an increase in test ordering. In fact, Isabel should help you drive more appropriate test ordering and, most likely, save money on unnecessary tests while helping to arrive at the correct diagnosis quicker.
On the contrary, Isabel is a tool that stimulates thinking about a patient’s presentation and assists with thoughtful consideration of the diagnoses being presented. Isabel does not tell you the diagnosis or tell you what to do next as it is designed to help broaden the differential.
Isabel is used when you have a patient where there is diagnostic uncertainty. For example, use Isabel on a patient seen in the ED who later returns with the same or worsening symptoms. Other use cases are a patient seen in the practice who is not responding to treatment, a patient admitted with a non-specific diagnosis, or a patient where something just doesn’t add up!
Yes, Isabel will get used if it’s properly implemented and part of your institutional drive to improve quality and diagnosis decision making. Isabel is simply a tool to help you improve your diagnostic process and help with your clinical reasoning.
Isabel is made available to clinicians where and when they want it. In addition to using Isabel standalone or as a mobile app, it has also been integrated with most of the leading EMRs such as Cerner, Epic, NextGen, T-System, SystemOne, NextGen, etc. Isabel integration options include everything from simply launching the application, to HL-7 InfoButton integration or fully embedded using the Isabel API (Application Programming Interface) suite. Isabel Active Intelligence uses the Isabel API and an NLP engine to extract the clinical features from existing free text documentation and provide a list of diagnoses without any additional data entry by the clinician. This is currently available with Cerner.
There are a number of differences that set Isabel apart from other tools. First, Isabel is built differently. It works by applying artificial intelligence pattern recognition software onto a hand crafted database of over 10,000 diagnosis presentations. Most other tools rely on a limited decision tree architecture. In addition, no other tool can boast Isabel’s combination of accuracy, speed and ease of use with the ability to handle complex clinical cases across all specialities while at the same time being integrated into the clinician’s workflow.
There is now an extensive body of validation studies supporting the use of Isabel. Recent RCT (Randomised Control Trial) studies show that Isabel substantially increases the diagnostic accuracy of clinicians.