Abstract
Objective: To test the effectiveness of a novel computer algorithm in generating correct differential diagnoses as a supporting tool in clinical decision-making
Background: Many computer based diagnostic algorithms to support clinical decision making have evolved over the past years. We (first two authors) have created a computer algorithm that has subsequently been made into an application (app) on iPhone platform (iOS). The algorithm generates a maximum of thirteen differential diagnoses in a descending order of probability, based on clinical and radiological inputs by the user. The algorithm resides on iPhone as a standalone application and does not need any external resource including Internet to compute differential diagnoses.
Design/Methods: We compared the differentials generated by the algorithm with clinical and pathological diagnoses in a series of autopsy/biopsy proven cases. Towards this end we used 100 previously published NEJM ‘Case records of the Massachusetts General Hospital’ and 30 unpublished Clinico-Pathologic Conference cases(CPC) from Postgraduate Institute of Medical Education and Research(PGIMER), Chandigarh, India. In each case record, we tabulated key symptoms, signs, MRI features, clinical discussant’s differentials and final diagnoses. The key symptoms, signs and MRI features were used as inputs for the algorithm. The differential diagnoses computed by the algorithm were then compared with the differentials of the clinical discussant and the final autopsy/biopsy diagnoses.
Results: Our algorithm correctly identified final diagnosis in 82% of NEJM case records and 89% of PGIMER CPCs. We then compared the differentials of clinical discussant with that of our algorithm. The algorithm, in 76% of NEJM cases and 82% in PGIMER CPCs, also generated the differentials put forward by the clinical discussant.
Conclusions: Smart algorithms coupled with a massive computing power, can be a powerful tool at the hands of a budding neurologist. Our app harnesses this concept on a readily available mobile platform (iOS) with satisfactory results.
Disclosure: Dr. Vinny has nothing to disclose. Dr. Vishnu has nothing to disclose. Dr. Balaini has nothing to disclose. Dr. Modi has nothing to disclose. Dr. Radotra has nothing to disclose. Dr. Lal has nothing to disclose.
Disputes & Debates: Rapid online correspondence
REQUIREMENTS
If you are uploading a letter concerning an article:
You must have updated your disclosures within six months: http://submit.neurology.org
Your co-authors must send a completed Publishing Agreement Form to Neurology Staff (not necessary for the lead/corresponding author as the form below will suffice) before you upload your comment.
If you are responding to a comment that was written about an article you originally authored:
You (and co-authors) do not need to fill out forms or check disclosures as author forms are still valid
and apply to letter.
Submission specifications:
- Submissions must be < 200 words with < 5 references. Reference 1 must be the article on which you are commenting.
- Submissions should not have more than 5 authors. (Exception: original author replies can include all original authors of the article)
- Submit only on articles published within 6 months of issue date.
- Do not be redundant. Read any comments already posted on the article prior to submission.
- Submitted comments are subject to editing and editor review prior to posting.