Current Volume 9
Abstract
A persistent gap exists between how clinicians actu- ally assess patients and how digital triage tools process symptom input. In any real consultation, a physician does not simply record whether a symptom is present—she probes its severity, asks how disruptive it has been, and weighs that rating against other findings. Consumer-facing symptom checkers, by contrast, have largely continued to rely on binary yes/no input since their inception. That design choice discards the one dimension of self-reported data that most reliably distinguishes between conditions with overlapping symptom sets. This paper introduces SymptomAI, a full-stack web application built to address that shortfall. Users rate each symptom on a 1–100 continuous slider rather than ticking a checkbox; those ratings drive a weighted scoring engine that evaluates candidate conditions against both symptom coverage and reported severity. After a primary diag- nosis is identified, a condition-specific follow-up stage presents three targeted questions, one per screen, and collects ternary (Yes/No/Unsure) responses that are merged into the final report. That report—carrying the diagnosis, confidence value, and five categories of clinical guidance—is compiled and exported as a PDF purely within the user’s browser. No health data returns to the server after the initial inference call. The technology stack comprises Next.js 14 (App Router), React 18, Clerk for federated authentication, MongoDB Atlas for the disease knowledge base, Tailwind CSS, and Framer Motion. When the scoring engine encounters two or more high-specificity symptoms rated simul- taneously above 75, a critical-marker escalation rule fires and boosts the confidence of the most likely condition. Validation on 100 simulated cases placed overall primary-diagnosis accuracy at 86 %, a margin of 22 percentage points over an equal-weight binary baseline. A parallel study with 30 participants showed that an “Intelligence Feed” sidebar naming the system’s active operations raised perceived trustworthiness by 41.4 % relative to a no-feed control.
Keywords
SymptomAI, Intensity Mapping, Differential Diagnosis, Tele-health, Next.js, MongoDB, Clerk Authentication, React, jsPDF, html2canvas, Weighted Scoring, Progressive Elicitation.
Citations
IRE Journals:
K. Nikhil Goud, Maniyam Bhanu Teja, M. Tanmay Reddy, Dr. Ramachandra, Shaik Meeravali "SymptomAI: A Dual-Phase Full-Stack Diagnostic Framework Combining Intensity-Weighted" Iconic Research And Engineering Journals Volume 9 Issue 10 2026 Page 3485-3494 https://doi.org/10.64388/IREV9I10-1716918
IEEE:
K. Nikhil Goud, Maniyam Bhanu Teja, M. Tanmay Reddy, Dr. Ramachandra, Shaik Meeravali
"SymptomAI: A Dual-Phase Full-Stack Diagnostic Framework Combining Intensity-Weighted" Iconic Research And Engineering Journals, 9(10) https://doi.org/10.64388/IREV9I10-1716918
About IRE Journals
IRE Journals is an open access online journals established with an aim to publish high quality of research work in various diciplines. We have more than 6 years in publications. Over 2500+ papers already published.
Important Links
For Authors
Volume 8:
Issue 1, Issue 2, Issue 3, Issue 4, Issue 5, Issue 6, Issue 7, Issue 8, Issue 9, Issue 10, Issue 11, Issue 12Volume 7:
Issue 1, Issue 2, Issue 3, Issue 4, Issue 5, Issue 6, Issue 7, Issue 8, Issue 9, Issue 10, Issue 11, Issue 12Volume 6:
Issue 1, Issue 2, Issue 3, Issue 4, Issue 5, Issue 6, Issue 7, Issue 8, Issue 9, Issue 10, Issue 11, Issue 12Volume 5:
Issue 1, Issue 2, Issue 3, Issue 4, Issue 5, Issue 6, Issue 7, Issue 8, Issue 9, Issue 10, Issue 11, Issue 12Volume 4:
Issue 1, Issue 2, Issue 3, Issue 4, Issue 5, Issue 6, Issue 7, Issue 8, Issue 9, Issue 10, Issue 11, Issue 12Volume 3:
Issue 1, Issue 2, Issue 3, Issue 4, Issue 5, Issue 6, Issue 7, Issue 8, Issue 9, Issue 10, Issue 11, Issue 12