Page 169 - {PDF Title}

Blockchain for secure e-health data in smart cities

The low sanitization values confirm that the proposed 8.6. Convergence analysis
STI-TSA with improved ARM incurs slight alteration Figure 5 shows the cost analysis using the proposed
of sensitive data over others. The high restoration STI-TSA with an improved ARM over STO, TSA,
values confirm the accurate retrieval of original data PFO, SBOA, BA, EHO-OBL, and BFL-PSO. To
34
35
using the STI-TSA with improved ARM over extant ensure better PP, the cost values should be low with
ones. fast convergence. This is well accomplished using the
proposed STI-TSA with an improved ARM approach.
Table 9. Data restoration analysis on privacy At initial iterations, the costs are high for all algorithms.
preservation for electronic health records with However, with increasing iterations, the costs are reduced.
blockchain technology Among all, the proposed STI-TSA with improved ARM
Methods 10% 20% 30% shows lower cost values and has a faster convergence rate
STO 0.850778 0.870158 0.892701 compared to other algorithms. Thus, the STI-TSA model
TSA 0.860621 0.881882 0.904179 could attain faster convergence and create high-quality
PFO 0.834467 0.864743 0.876605 solutions by including diverse optimizing approaches.
SBOA 0.822115 0.835485 0.862898 9. Conclusion
BA 0.814612 0.823373 0.847861
EHO-OBL 0.809184 0.824774 0.86858 This work presented a novel PP approach for EHR
BFL-PSO 0.826664 0.82739 0.886539 utilizing BT. The method used an organized procedure
STI-TSA 0.933553 0.945149 0.953422 that included data sanitization and restoration. First,
improved ARM was used to detect sensitive information.
Abbreviations: BA: Bat Algorithm; BFL-PSO: Bee-foraging
learning particle swarm optimization; EHO-OBL: Elephant The STI-TSA then finds the best key to improve data
Herding Optimization with Opposition-based Learning; PFO: security. Sensitive information was safeguarded by
Puffer Fish Optimization; SBOA: Secretary Bird Optimization applying an XOR operation among the sensitive
Algorithm; STO: Siberian Tiger Optimization; TSA: Tuna Swarm data and the optimum key to sanitize the data. Next,
algorithm. blockchain storage was used to store the sanitized data.
When necessary, the restoration procedure undoes the
XOR operation to retrieve the original sensitive data.
In the end, the reverse procedure of improved ARM
was used to recover the original health data. From the
analysis, when data were 10%, the proposed STI-TSA-
based optimization attained a high IPR of 0.93%. In
comparison, IPR of 0.94% and 0.95% were achieved
when data were at 20% and 30%, respectively. At 30%
of data, extant optimization schemes, such as STO,
TSA, PFO, SBOA, BA, EHO-OBL, and BFL-PSO,
attained low IPR of 0.89, 0.9, 0.87, 0.9, 0.87, 0.89, and
0.88, respectively. In the future, DL approaches can be
integrated to improve the privacy of EHR data.

Acknowledgments

Figure 5. Convergence study of privacy preservation for None.
electronic health records using blockchain technology.
Abbreviations: BA: Bat Algorithm; BFL-PSO: Bee- Funding
foraging learning particle swarm optimization;
EHO-OBL: Elephant Herding Optimization with None.
Opposition-based Learning; PFO: Puffer Fish
Optimization; SBOA: Secretary Bird Optimization Conflict of interest
Algorithm; STO: Siberian Tiger Optimization; TSA:
Tuna Swarm algorithm. The authors declare no conflicts of interest.

Volume 22 Issue 1 (2025) 163 doi: 10.36922/AJWEP025040017