Machine Learning for Lung Cancer Subtype Classification: Combining Clinical, Histopathological, and Biophysical Features
<b>Background/Objectives:</b> Despite advances in diagnostic techniques, accurate classification of lung cancer subtypes remains crucial for treatment planning. Traditional methods like genomic studies face limitations such as high cost and complexity. This study investigates whether int...
Saved in:
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2025-01-01
|
| Series: | Diagnostics |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2075-4418/15/2/127 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1832588671285460992 |
|---|---|
| author | Aiga Andrijanova Lasma Bugovecka Sergejs Isajevs Donats Erts Uldis Malinovskis Andis Liepins |
| author_facet | Aiga Andrijanova Lasma Bugovecka Sergejs Isajevs Donats Erts Uldis Malinovskis Andis Liepins |
| author_sort | Aiga Andrijanova |
| collection | DOAJ |
| description | <b>Background/Objectives:</b> Despite advances in diagnostic techniques, accurate classification of lung cancer subtypes remains crucial for treatment planning. Traditional methods like genomic studies face limitations such as high cost and complexity. This study investigates whether integrating atomic force microscopy (AFM) measurements with conventional clinical and histopathological data can improve lung cancer subtype classification. <b>Methods:</b> We developed and analyzed a novel dataset combining clinical, histopathological, and AFM-derived biophysical characteristics from 37 lung cancer patients. Various machine learning techniques were evaluated, with a focus on Bayesian Networks due to their ability to handle complex data with missing values. Leave-One-Out Cross-Validation was employed to assess model performance. <b>Results:</b> The integration of biophysical features improved classification accuracy from 86.49% to 89.19% using a data-driven Bayesian Network model, though this improvement was not statistically significant (<i>p</i> = 1.0). Four key features were identified as highly predictive: sex, vascular invasion, perineural invasion, and ALK mutation. A simplified model using only these features achieved identical performance with significantly reduced complexity (BIC 51.931 vs. 268.586). <b>Conclusions:</b> While AFM-derived measurements showed promise for enhancing lung cancer subtype classification, larger datasets are needed to fully validate their impact. Our findings demonstrate the feasibility of incorporating biophysical measurements into cancer classification frameworks and identify the most predictive features for accurate diagnosis. Further research with expanded datasets is needed to validate these findings. |
| format | Article |
| id | doaj-art-0d308d5d5faa4983834e653dccfb6580 |
| institution | Kabale University |
| issn | 2075-4418 |
| language | English |
| publishDate | 2025-01-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Diagnostics |
| spelling | doaj-art-0d308d5d5faa4983834e653dccfb65802025-01-24T13:28:49ZengMDPI AGDiagnostics2075-44182025-01-0115212710.3390/diagnostics15020127Machine Learning for Lung Cancer Subtype Classification: Combining Clinical, Histopathological, and Biophysical FeaturesAiga Andrijanova0Lasma Bugovecka1Sergejs Isajevs2Donats Erts3Uldis Malinovskis4Andis Liepins5SIA “APPLY”, Ieriku Street 5, LV-1084 Riga, LatviaInstitute of Chemical Physics, Faculty of Science and Technology, University of Latvia, Jelgavas Street 1, LV-1004 Riga, LatviaFaculty of Medicine and Life Sciences, University of Latvia, Jelgavas Street 3, LV-1004 Riga, LatviaInstitute of Chemical Physics, Faculty of Science and Technology, University of Latvia, Jelgavas Street 1, LV-1004 Riga, LatviaInstitute of Chemical Physics, Faculty of Science and Technology, University of Latvia, Jelgavas Street 1, LV-1004 Riga, LatviaSIA “APPLY”, Ieriku Street 5, LV-1084 Riga, Latvia<b>Background/Objectives:</b> Despite advances in diagnostic techniques, accurate classification of lung cancer subtypes remains crucial for treatment planning. Traditional methods like genomic studies face limitations such as high cost and complexity. This study investigates whether integrating atomic force microscopy (AFM) measurements with conventional clinical and histopathological data can improve lung cancer subtype classification. <b>Methods:</b> We developed and analyzed a novel dataset combining clinical, histopathological, and AFM-derived biophysical characteristics from 37 lung cancer patients. Various machine learning techniques were evaluated, with a focus on Bayesian Networks due to their ability to handle complex data with missing values. Leave-One-Out Cross-Validation was employed to assess model performance. <b>Results:</b> The integration of biophysical features improved classification accuracy from 86.49% to 89.19% using a data-driven Bayesian Network model, though this improvement was not statistically significant (<i>p</i> = 1.0). Four key features were identified as highly predictive: sex, vascular invasion, perineural invasion, and ALK mutation. A simplified model using only these features achieved identical performance with significantly reduced complexity (BIC 51.931 vs. 268.586). <b>Conclusions:</b> While AFM-derived measurements showed promise for enhancing lung cancer subtype classification, larger datasets are needed to fully validate their impact. Our findings demonstrate the feasibility of incorporating biophysical measurements into cancer classification frameworks and identify the most predictive features for accurate diagnosis. Further research with expanded datasets is needed to validate these findings.https://www.mdpi.com/2075-4418/15/2/127lung cancercancer subtype classificationatomic force microscopybiophysical propertiesmachine learningBayesian networks |
| spellingShingle | Aiga Andrijanova Lasma Bugovecka Sergejs Isajevs Donats Erts Uldis Malinovskis Andis Liepins Machine Learning for Lung Cancer Subtype Classification: Combining Clinical, Histopathological, and Biophysical Features Diagnostics lung cancer cancer subtype classification atomic force microscopy biophysical properties machine learning Bayesian networks |
| title | Machine Learning for Lung Cancer Subtype Classification: Combining Clinical, Histopathological, and Biophysical Features |
| title_full | Machine Learning for Lung Cancer Subtype Classification: Combining Clinical, Histopathological, and Biophysical Features |
| title_fullStr | Machine Learning for Lung Cancer Subtype Classification: Combining Clinical, Histopathological, and Biophysical Features |
| title_full_unstemmed | Machine Learning for Lung Cancer Subtype Classification: Combining Clinical, Histopathological, and Biophysical Features |
| title_short | Machine Learning for Lung Cancer Subtype Classification: Combining Clinical, Histopathological, and Biophysical Features |
| title_sort | machine learning for lung cancer subtype classification combining clinical histopathological and biophysical features |
| topic | lung cancer cancer subtype classification atomic force microscopy biophysical properties machine learning Bayesian networks |
| url | https://www.mdpi.com/2075-4418/15/2/127 |
| work_keys_str_mv | AT aigaandrijanova machinelearningforlungcancersubtypeclassificationcombiningclinicalhistopathologicalandbiophysicalfeatures AT lasmabugovecka machinelearningforlungcancersubtypeclassificationcombiningclinicalhistopathologicalandbiophysicalfeatures AT sergejsisajevs machinelearningforlungcancersubtypeclassificationcombiningclinicalhistopathologicalandbiophysicalfeatures AT donatserts machinelearningforlungcancersubtypeclassificationcombiningclinicalhistopathologicalandbiophysicalfeatures AT uldismalinovskis machinelearningforlungcancersubtypeclassificationcombiningclinicalhistopathologicalandbiophysicalfeatures AT andisliepins machinelearningforlungcancersubtypeclassificationcombiningclinicalhistopathologicalandbiophysicalfeatures |