BECF: A Unified Cross-Layer Risk Scoring Framework for Healthcare IoT Security with Blockchain Anomaly Detection and Explainable Intrusion Detection

Healthcare Internet of Medical Things (IoMT) security and blockchain consent governance remain siloed: intrusion detection systems monitor network traffic but ignore blockchain transaction anomalies, while blockchain anomaly detectors largely target general financial contexts. We propose the Blockchain-Enabled Cybersecurity Framework (BECF), a benchmark-validated architectural proof-of-concept that integrates temporally validated, SHAP-explainable gradient boosting for intrusion detection on ToN-IoT, probability-calibrated blockchain anomaly detection on Elliptic, smart-contract consent management on Ethereum Sepolia, and a composite risk function fusing network intrusion probability, blockchain anomaly probability, and patient data sensitivity. Fusion weights and classifiers are optimized through constrained grid search and Bayesian optimization, with bootstrap confidence intervals and controlled post-hoc comparisons. On ToN-IoT, the Bayesian-optimized LightGBM classifier achieves 99.89% accuracy (95% CI: [99.85%, 99.92%]), F1 = 0.9993, and MCC = 0.9969; on Elliptic, LightGBM achieves 76.63% balanced accuracy, F1 = 0.6290, and AUPRC = 0.5854 for general illicit Bitcoin transaction detection—not healthcare-consent attack detection. Smart contracts achieve 99.8% success, zero critical findings from the applied analyzers, and approximately 17.8% gas savings. Across 10,000 controlled integration-test events, BECF achieves a 74.4% event-level true positive rate for dual-layer alerts with a false-positive rate of 0.20%, and detects all 47 explicitly constructed coordinated scenarios (47/47; 100%; 95% Wilson CI: [92.4%, 100%]). These scenarios were generated in a controlled integration testbed and do not represent naturally occurring clinical attacks; therefore, the scenario result is controlled feasibility evidence rather than clinical detection performance. Prospective hospital validation and Ethereum-native attack validation remain necessary before clinical deployment.

Authors

Institutions

Publication Details

Journal
Computers
Published
2026-11-20
DOI
https://doi.org/10.3390/computers15110861
Primary Topic
Blockchain Technology Applications and Security
Type
article
Field-Weighted Citation Impact
0.00
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
NOV
article

BECF: A Unified Cross-Layer Risk Scoring Framework for Healthcare IoT Security with Blockchain Anomaly Detection and Explainable Intrusion Detection

Muhammad Ismail Mohmand, Usama Ahmed, Humaira Arshad, Usman Ahmad et al.
1 links
Computers
Blockchain Technology Applications and Security
article

BECF: A Unified Cross-Layer Risk Scoring Framework for Healthcare IoT Security with Blockchain Anomaly Detection and Explainable Intrusion Detection

Muhammad Ismail Mohmand, Usama Ahmed, Humaira Arshad, Usman Ahmad, Usman Mohyud Din Chaudhary
article en

Abstract

Healthcare Internet of Medical Things (IoMT) security and blockchain consent governance remain siloed: intrusion detection systems monitor network traffic but ignore blockchain transaction anomalies, while blockchain anomaly detectors largely target general financial contexts. We propose the Blockchain-Enabled Cybersecurity Framework (BECF), a benchmark-validated architectural proof-of-concept that integrates temporally validated, SHAP-explainable gradient boosting for intrusion detection on ToN-IoT, probability-calibrated blockchain anomaly detection on Elliptic, smart-contract consent management on Ethereum Sepolia, and a composite risk function fusing network intrusion probability, blockchain anomaly probability, and patient data sensitivity. Fusion weights and classifiers are optimized through constrained grid search and Bayesian optimization, with bootstrap confidence intervals and controlled post-hoc comparisons. On ToN-IoT, the Bayesian-optimized LightGBM classifier achieves 99.89% accuracy (95% CI: [99.85%, 99.92%]), F1 = 0.9993, and MCC = 0.9969; on Elliptic, LightGBM achieves 76.63% balanced accuracy, F1 = 0.6290, and AUPRC = 0.5854 for general illicit Bitcoin transaction detection—not healthcare-consent attack detection. Smart contracts achieve 99.8% success, zero critical findings from the applied analyzers, and approximately 17.8% gas savings. Across 10,000 controlled integration-test events, BECF achieves a 74.4% event-level true positive rate for dual-layer alerts with a false-positive rate of 0.20%, and detects all 47 explicitly constructed coordinated scenarios (47/47; 100%; 95% Wilson CI: [92.4%, 100%]). These scenarios were generated in a controlled integration testbed and do not represent naturally occurring clinical attacks; therefore, the scenario result is controlled feasibility evidence rather than clinical detection performance. Prospective hospital validation and Ethereum-native attack validation remain necessary before clinical deployment.

ComputersVol. 15(11)
University of Lahore (PK), Islamia University of Bahawalpur (PK), National University of Computer and Emerging Sciences (PK), Atlas Üniversitesi (TR), Bath Spa University (GB)
Good health and well-being
Blockchain Technology Applications and Security
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.

Laika access coming soon