Machine Learning Diffuse Optical Tomography Using Extreme Gradient Boosting and Genetic Programming
Diffuse optical tomography (DOT) is a non-invasive method for detecting breast cancer; however, it struggles to produce high-quality images due to the complexity of scattered light and the limitations of traditional image reconstruction algorithms. These algorithms can be affected by boundary condit...
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2023-03-01
|
| Series: | Bioengineering |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2306-5354/10/3/382 |
| _version_ | 1797613401422168064 |
|---|---|
| author | Ami Hauptman Ganesh M. Balasubramaniam Shlomi Arnon |
| author_facet | Ami Hauptman Ganesh M. Balasubramaniam Shlomi Arnon |
| author_sort | Ami Hauptman |
| collection | DOAJ |
| description | Diffuse optical tomography (DOT) is a non-invasive method for detecting breast cancer; however, it struggles to produce high-quality images due to the complexity of scattered light and the limitations of traditional image reconstruction algorithms. These algorithms can be affected by boundary conditions and have a low imaging accuracy, a shallow imaging depth, a long computation time, and a high signal-to-noise ratio. However, machine learning can potentially improve the performance of DOT by being better equipped to solve inverse problems, perform regression, classify medical images, and reconstruct biomedical images. In this study, we utilized a machine learning model called “XGBoost” to detect tumors in inhomogeneous breasts and applied a post-processing technique based on genetic programming to improve accuracy. The proposed algorithm was tested using simulated DOT measurements from complex inhomogeneous breasts and evaluated using the cosine similarity metrics and root mean square error loss. The results showed that the use of XGBoost and genetic programming in DOT could lead to more accurate and non-invasive detection of tumors in inhomogeneous breasts compared to traditional methods, with the reconstructed breasts having an average cosine similarity of more than 0.97 ± 0.07 and average root mean square error of around 0.1270 ± 0.0031 compared to the ground truth. |
| first_indexed | 2024-03-11T06:55:18Z |
| format | Article |
| id | doaj.art-3af99c721c144c3caf4071ae7c2ace85 |
| institution | Directory Open Access Journal |
| issn | 2306-5354 |
| language | English |
| last_indexed | 2024-03-11T06:55:18Z |
| publishDate | 2023-03-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Bioengineering |
| spelling | doaj.art-3af99c721c144c3caf4071ae7c2ace852023-11-17T09:40:31ZengMDPI AGBioengineering2306-53542023-03-0110338210.3390/bioengineering10030382Machine Learning Diffuse Optical Tomography Using Extreme Gradient Boosting and Genetic ProgrammingAmi Hauptman0Ganesh M. Balasubramaniam1Shlomi Arnon2Department of Computer Science, Sapir Academic College, Sderot 7915600, IsraelDepartment of Electrical and Computer Engineering, Ben-Gurion University of the Negev, Be’er Sheva 8441405, IsraelDepartment of Electrical and Computer Engineering, Ben-Gurion University of the Negev, Be’er Sheva 8441405, IsraelDiffuse optical tomography (DOT) is a non-invasive method for detecting breast cancer; however, it struggles to produce high-quality images due to the complexity of scattered light and the limitations of traditional image reconstruction algorithms. These algorithms can be affected by boundary conditions and have a low imaging accuracy, a shallow imaging depth, a long computation time, and a high signal-to-noise ratio. However, machine learning can potentially improve the performance of DOT by being better equipped to solve inverse problems, perform regression, classify medical images, and reconstruct biomedical images. In this study, we utilized a machine learning model called “XGBoost” to detect tumors in inhomogeneous breasts and applied a post-processing technique based on genetic programming to improve accuracy. The proposed algorithm was tested using simulated DOT measurements from complex inhomogeneous breasts and evaluated using the cosine similarity metrics and root mean square error loss. The results showed that the use of XGBoost and genetic programming in DOT could lead to more accurate and non-invasive detection of tumors in inhomogeneous breasts compared to traditional methods, with the reconstructed breasts having an average cosine similarity of more than 0.97 ± 0.07 and average root mean square error of around 0.1270 ± 0.0031 compared to the ground truth.https://www.mdpi.com/2306-5354/10/3/382diffuse optical tomographyextreme gradient boostinggenetic programminginhomogeneous breastinverse problems |
| spellingShingle | Ami Hauptman Ganesh M. Balasubramaniam Shlomi Arnon Machine Learning Diffuse Optical Tomography Using Extreme Gradient Boosting and Genetic Programming Bioengineering diffuse optical tomography extreme gradient boosting genetic programming inhomogeneous breast inverse problems |
| title | Machine Learning Diffuse Optical Tomography Using Extreme Gradient Boosting and Genetic Programming |
| title_full | Machine Learning Diffuse Optical Tomography Using Extreme Gradient Boosting and Genetic Programming |
| title_fullStr | Machine Learning Diffuse Optical Tomography Using Extreme Gradient Boosting and Genetic Programming |
| title_full_unstemmed | Machine Learning Diffuse Optical Tomography Using Extreme Gradient Boosting and Genetic Programming |
| title_short | Machine Learning Diffuse Optical Tomography Using Extreme Gradient Boosting and Genetic Programming |
| title_sort | machine learning diffuse optical tomography using extreme gradient boosting and genetic programming |
| topic | diffuse optical tomography extreme gradient boosting genetic programming inhomogeneous breast inverse problems |
| url | https://www.mdpi.com/2306-5354/10/3/382 |
| work_keys_str_mv | AT amihauptman machinelearningdiffuseopticaltomographyusingextremegradientboostingandgeneticprogramming AT ganeshmbalasubramaniam machinelearningdiffuseopticaltomographyusingextremegradientboostingandgeneticprogramming AT shlomiarnon machinelearningdiffuseopticaltomographyusingextremegradientboostingandgeneticprogramming |