An Enhanced Measurement of Epicardial Fat Segmentation and Severity Classification using Modified U-Net and FOA-guided XGBoost

K Rajalakshmi; S Palanivel Rajan

doi:10.2478/msr-2025-0012

.blurhash-client-img { display: none !important; }

An Enhanced Measurement of Epicardial Fat Segmentation and Severity Classification using Modified U-Net and FOA-guided XGBoost

Measurement Science Review

Volume 25 (2025): Issue 2 (April 2025)

By: K Rajalakshmi and S Palanivel Rajan

Open Access

|Jun 2025

Figures & Tables

Pseudocode:

FOA Feature Selection and XGBoost Optimization

# *Initialize XGBoost parameters* xgb_params = { 'learning_rate': 0.1, 'max_depth': 5, 'n_estimators': 100, # *Other parameters...* } # *Feature selection using FOA* def FOA_feature_selection(feature_set, labels): # *Implement FOA for feature selection* # *Initialize population* initialize_population() # *Evaluate fitness of each individual* evaluate_fitness(feature_set, labels) # *Repeat iterations* while not stopping_criteria_met(): # *Apply selection, crossover, and mutation* apply_selection_crossover_mutation() # *Evaluate fitness of new individuals* evaluate_fitness(feature_set, labels) # *Return selected features from the best individual* return best_individual_features # *Optimize XGBoost parameters with FOA* def FOA_optimize_xgb(xgb_params, selected_features, labels): best_params = xgb_params.copy() # *Define FOA iterations for parameter optimization* for iteration in range(max_iterations): # *Perform optimization on XGBoost parameters using FOA* for param in xgb_params: # *Set range for the parameter* param_range = define_parameter_range(param) # *Apply FOA to optimize the parameter* updated_param_value = optimize_parameter(param_range) # *Update XGBoost parameters with optimized value* best_params[param] = updated_param_value # *Train XGBoost model with updated parameters* trained_model = train_xgboost_model(best_params, selected_features, labels) # *Evaluate model performance and update best_params if needed* if model_performance_improved(trained_model): best_params = updated_params return best_params # *Run feature selection and XGBoost optimization* selected_features = FOA_feature_selection(feature_set, labels) best_xgb_params = FOA_optimize_xgb(xgb_params, selected_features, labels

Segmentation result (epicardial and mediastinal fats).

Training and validation loss of the proposed model over epochs.

Table 1.

Performance analysis.

Model	Mean IOU [%]	MDS [%]	PCC
FCN	68.92	73.5	0.789
U-Net	71.5	79.7	0.8721
Seg-Net	76.7	85.42	0.8905
Attention U-Net	80.2	88.54	0.932
SAR-U-Net	86.78	92.6	0.954
MSE-GRU-U-Net	89.5	94.3	0.973

Table 2.

Performance analysis.

Model	FOA-ID3 [%]	FOA-NB [%]	FOA-XGBoost [%]
Accuracy	92	95	97
Specificity	91	95	96
Precision	90	94	97
Recall	94	94	98

References

Authors

Metrics

Articles in this issue

DOI: https://doi.org/10.2478/msr-2025-0012 | Journal eISSN: 1335-8871

Journal RSS Feed

Language: English

Page range: 93 - 99

Submitted on: Jun 23, 2024

Accepted on: Apr 15, 2025

Published on: Jun 7, 2025

Published by: Slovak Academy of Sciences, Institute of Measurement Science

In partnership with: Paradigm Publishing Services

Publication frequency: Volume open

Keywords:

epicardial fat segmentation,

U-Net architecture,

falcon optimization algorithm,

feature selection,

severity classification

Related subjects:

Engineering,

Electrical engineering,

Control engineering, metrology and testing

© 2025 K Rajalakshmi, S Palanivel Rajan, published by Slovak Academy of Sciences, Institute of Measurement Science
This work is licensed under the Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 License.

Volume 25 (2025): Issue 2 (April 2025)

An Enhanced Measurement of Epicardial Fat Segmentation and Severity Classification using Modified U-Net and FOA-guided XGBoost

Figures & Tables

Fig. 1.

Fig. 2.

Pseudocode:

Fig. 3.

Fig. 4.

Fig. 5.

Table 1.

Table 2.

Paradigm

My account