Have a personal or library account? Click to login
Optimized Deep learning Frameworks for the Medical Image Transmission in IoMT Environment Cover

Optimized Deep learning Frameworks for the Medical Image Transmission in IoMT Environment

By: Rashmi P and  R. Gomathi  
Open Access
|Feb 2025

Figures & Tables

Figure 1:

Architecture for the recommended approach.
Architecture for the recommended approach.

Figure 2 :

Architecture for CNN
Architecture for CNN

Figure 3:

Architecture for LSTM
Architecture for LSTM

Figure 4:

Comparative analysis with the Residing approach
Comparative analysis with the Residing approach

Figure 5:

Convergence Analysis for the Proposed approach
Convergence Analysis for the Proposed approach

Performance metrics for the recommended approach

AlgorithmsPerformance Metrics (%)
AccuracyPrecisionRecallSpecificityF1-score
ANN84.386.386.586.585.3
CNN87.287.286.686.383.2
DNN88.588.184.388.686.3
RNN89.990.189.7488.388.2
Proposed Model98.397.097.896.296.0

Performance Comparison between the recommended architecture and Traditional Model

MetricProposed FrameworkExisting Method (Baseline)
Compression Ratio4:13:1
Encryption Time (ms)12 ms18 ms
PSNR (dB)38.534.2
Accuracy (Diagnosis)98%95%
Bandwidth Savings (%)75%60%

Performance measures utilized in the examination

SL.NOPerformance MeasuresExpression
1AccuracyTP+TNTP+TN+FP+FN
2RecallTPTP+FN×100
3SpecificityTNTN+FP
4PrecisionTNTP+FP
5F1-Score2.Precison*RecallPrecision+Recall
Language: English
Page range: 148 - 165
Submitted on: Sep 20, 2024
Accepted on: Oct 28, 2024
Published on: Feb 24, 2025
Published by: Future Sciences For Digital Publishing
In partnership with: Paradigm Publishing Services
Publication frequency: 2 issues per year

© 2025 Rashmi P, R. Gomathi, published by Future Sciences For Digital Publishing
This work is licensed under the Creative Commons Attribution 4.0 License.