ChatPPG.
Join free
First page or project page of “Patient-specific conformal calibration for arrhythmia monitoring: contamination, reviewer dependence and distribution shift” (opens in a new tab)

Paperdoi:10.5281/zenodo.23557385

Patient-specific conformal calibration for arrhythmia monitoring: contamination, reviewer dependence and distribution shift

Venue listed2026Added Oct 22, 2026

  • LT
  • SE
  • AU
  • T
  • AA

Authors: Lamia Fatiha Kazi Tani, Souaad Hamza Cherif (Biomedical Engineering, Abou Bekr Belkaid University, Tlemcen, Algeria); Mohammed Yassine Kazi Tani (ESI Sidi Bel Abbès)

Code + per-record results release (Zenodo 10.5281/zenodo.23557385, v1.0.0, 2026-10-22); manuscript 'Submitted to Expert Systems with Applications' per README (not public)

Publisher (opens in a new tab) Cite

Open codeLicence: manuscript not public; code MIT

First page or project page of “Patient-specific conformal calibration for arrhythmia monitoring: contamination, reviewer dependence and distribution shift” (opens in a new tab)

TL;DR

Beat-level ECG and PPG features → population random forests (ECG-RF, PPG-RF, early-fusion Joint-RF; late/SQI-gated/Bonferroni fusions also tested) plus a 1-D CNN baseline. Each patient's alarm threshold is then set by per-patient conformal p-values computed against a sliding buffer of adjudicated (reviewer-confirmed) beats (buffer M = 300, gap G = 20, α = 1 %, 5 % start-up review), with a CUSUM episode alarm on top.

  • Form factor: clinical_finger + wrist (external test)

Abstract

Code and per-record results for the paper 'Patient-specific conformal calibration for arrhythmia monitoring: contamination, reviewer dependence and distribution shift'. No patient data are included; the VitalDB Arrhythmia Database (PhysioNet) and the Vilnius database (Zenodo 10.5281/zenodo.23187681) must be obtained from their providers.

Verbatim from DataCite (opens in a new tab) · Licence: https://opensource.org/licenses/MIT

Results on ChatPPG leaderboards

All leaderboards

No results from this paper on ChatPPG leaderboards yet. Browse the leaderboards.

Tasks & signals

Tasks

AF / rhythm

Signals

PPGECG

Devices

Bittium Faros

Keywords

afibarrhythmiaconformal-predictionfalse-alarm-controlpersonalizationdistribution-shiftfinger-to-wrist-transferVitalDBVilniusECG-PPG-fusionrandom-forestreproducibility

Datasets

How it works

How Method works: Setup → Method → TargetSetup: Signals (PPG, ECG); Form factor (clinical_finger + wrist (external test)). Method: Method. Target: AF / rhythm.SetupSignals: PPG, ECGSignalsPPG, ECGForm factor: clinical_finger + wrist (external test)Form factorclinical_finger + wrist (external test)MethodMethodTargetAF / rhythmAF / rhythm

Cite this paper

APA

Tani, L. F. K., Cherif, S. H., & Tani, M. Y. K. (2026). Patient-specific conformal calibration for arrhythmia monitoring: contamination, reviewer dependence and distribution shift. Code + per-record results release. https://doi.org/10.5281/zenodo.23557385

BibTeX

@article{tani2026patientspecific,
  title         = {{Patient-specific conformal calibration for arrhythmia monitoring: contamination, reviewer dependence and distribution shift}},
  author        = {Tani, Lamia Fatiha Kazi and Cherif, Souaad Hamza and Tani, Mohammed Yassine Kazi},
  year          = {2026},
  journal       = {Code + per-record results release},
  doi           = {10.5281/zenodo.23557385}
}

Built from the hub record. Check it against the publisher before submitting.

Added to the ChatPPG research hub

Same tasks, data and authors

Papers

Models

Datasets