Table 2.

Comparative analysis of IoMT security schemes (centralized methods).

Author	Year	Objective	Technique Used	Type of Data	Framework	Pros	Cons
Deebak et al. [84]	2020	Data security and anonymity	PKI	Medical records	SSA	Solves Chiou et al.’s work [83]	Computational cost is high
Park et al. [85]	2020	To solve issues of MAKA scheme	PKI	Medical IoT data	LAKS-NVT	Does not require a server verification table	Traceable
Kumar et al. [87]	2020	Secure and efficient cloud-centric IoMT-enabled smart healthcare system	PKI	PHI file	EF-IDASC	Low energy consumption	DoS, reply attack
Limaye et al. [88]	2018	Facilitate research into new microarchitectures and optimizations	PKI	Healthcare data	HERMIT	Efficient processors for IoMT applications	Basic security
Lu et al. [89]	2020	TPM deployed in non-TPM protected embedded device via network	PKI	Sensor data	xTSeH	Does not discard request due to increased traffic	Security improvement required
Hsu et al. [90]	2020	Remove storing credentials and secure communication	PKI	eHealth data	UCSSO	No storage and central authority	Service could be interrupted
Chen et al. [91]	2021	Reduce energy consumption, achieve privacy and security	PKI & Chaotic map	Health data	-	Group authentication	Server impersonation
Li et al. [93]	2021	Reduce complexity and secure communication	PKI	Medical data	PSL-MAAKA	Lightweight scheme	Much time and storage required
Zhang et al. [94]	2020	Protect personal health records	ABE	PHR file	PHR sharing framework	Support offline and online	MITM, DoS, etc., security
Liu et al. [97]	2018	Enhance privacy preserving and efficient data structure	CP-ABE	Biomedical data	-	Server impersonation attack	Lot of storage and computation
Hwang et al. [95]	2020	Improve CP-ABE based scheme	CP-ABE	PHI file	-	Resolves key abuse problem	PHI leakage
Huang et al. [98]	2019	Protection from unauthorized entity	ECG	PHR file	-	Remove noise, light algorithm	No anonymous identity
Xu et al. [9]	2019	Secure data sharing	MAC	PHI file	-	Multi-keyword search	Device to gateway security
Siddiqi et al. [99]	2020	Security protocol for IMD ecosystem	MAC	Medical data	IMDfence	7% energy consumption	No user anonymity
Hahn et al. [96]	2020	Attack MAC-based scheme and countermeasure	Commitment (MAC)	Medical data	-	Low verification time	DoS, server impersonation
Li et al. [103]	2019	Enhance security of previous work	ECC	Medical data	3FUAP	Vulnerability and countermeasure	Computational cost
Almog- ren et al. [104]	2020	Fake node detection and deactivation	ECC	eHealth data	FTM	Double filter	Mainly focused on Sybil attack
Ying et al. [105]	2021	Secure communication	ECC	Medical data	-	Low computational time	High communication overhead
Liu et al. [106]	2021	Achieve data SNP preservation	ECC	EHR file	-	Major decryption on server side	Complex
Wang et al. [107]	2020	Ensure data privacy	Machine learning	Medical data	EPoSVM	Efficiency	Significant time required
Awan et al. [108]	2020	Maintains a robust network by predicting and eliminating malicious nodes	Supervised learning and ECC	Health data	NeuroTrust	Lightweight encryption	Needs focus on attacks
Ding et al. [109]	2020	To preserve the privacy or security of the patient	Deep learning	DeepEDN	Image	Fast	Needs robustness and server verification
Yanambaka et al. [111]	2019	Secure communication	PUF	Medical data	Pmsec	Lightweight	ML attack
Gope et al. [112]	2020	Secure and efficient authentication	PUF	Healthcare monitoring	-	Less computation at server	Two CRPs per transaction
Alladi et al. [78]	2020	To achieve physical security	PUF	Health data	HARCI	Low time in computation	Unstable CRP can cause failure