EMM-LC Fusion: Enhanced Multimodal Fusion for Lung Cancer Classification
Lung cancer (LC) is the most common cause of cancer-related deaths in the UK due to delayed diagnosis. The existing literature establishes a variety of factors which contribute to this, including the misjudgement of anatomical structure by doctors and radiologists. This study set out to develop a so...
| Main Authors: | , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2022-08-01
|
| Series: | AI |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2673-2688/3/3/38 |
| _version_ | 1797492073479274496 |
|---|---|
| author | James Barrett Thiago Viana |
| author_facet | James Barrett Thiago Viana |
| author_sort | James Barrett |
| collection | DOAJ |
| description | Lung cancer (LC) is the most common cause of cancer-related deaths in the UK due to delayed diagnosis. The existing literature establishes a variety of factors which contribute to this, including the misjudgement of anatomical structure by doctors and radiologists. This study set out to develop a solution which utilises multiple modalities in order to detect the presence of LC. A review of the existing literature established failings within methods to exploit rich intermediate feature representations, such that it can capture complex multimodal associations between heterogenous data sources. The methodological approach involved the development of a novel machine learning (ML) model to facilitate quantitative analysis. The proposed solution, named EMM-LC Fusion, extracts intermediate features from a pre-trained modified AlignedXception model and concatenates these with linearly inflated features of Clinical Data Elements (CDE). The implementation was evaluated and compared against existing literature using F1 score, average precision (AP), and area under curve (AUC) as metrics. The findings presented in this study show a statistically significant improvement (<i>p</i> < 0.05) upon the previous fusion method, with an increase in F-Score from 0.402 to 0.508. The significance of this establishes that the extraction of intermediate features produces a fertile environment for the detection of intermodal relationships for the task of LC classification. This research also provides an architecture to facilitate the future implementation of alternative biomarkers for lung cancer, one of the acknowledged limitations of this study. |
| first_indexed | 2024-03-10T00:57:25Z |
| format | Article |
| id | doaj.art-2135bdf6689b4526911fb08daa348088 |
| institution | Directory Open Access Journal |
| issn | 2673-2688 |
| language | English |
| last_indexed | 2024-03-10T00:57:25Z |
| publishDate | 2022-08-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | AI |
| spelling | doaj.art-2135bdf6689b4526911fb08daa3480882023-11-23T14:39:45ZengMDPI AGAI2673-26882022-08-013365968210.3390/ai3030038EMM-LC Fusion: Enhanced Multimodal Fusion for Lung Cancer ClassificationJames Barrett0Thiago Viana1Cyber and Technical Computing, University of Gloucestershire, Cheltenham GL50 2RH, UKCyber and Technical Computing, University of Gloucestershire, Cheltenham GL50 2RH, UKLung cancer (LC) is the most common cause of cancer-related deaths in the UK due to delayed diagnosis. The existing literature establishes a variety of factors which contribute to this, including the misjudgement of anatomical structure by doctors and radiologists. This study set out to develop a solution which utilises multiple modalities in order to detect the presence of LC. A review of the existing literature established failings within methods to exploit rich intermediate feature representations, such that it can capture complex multimodal associations between heterogenous data sources. The methodological approach involved the development of a novel machine learning (ML) model to facilitate quantitative analysis. The proposed solution, named EMM-LC Fusion, extracts intermediate features from a pre-trained modified AlignedXception model and concatenates these with linearly inflated features of Clinical Data Elements (CDE). The implementation was evaluated and compared against existing literature using F1 score, average precision (AP), and area under curve (AUC) as metrics. The findings presented in this study show a statistically significant improvement (<i>p</i> < 0.05) upon the previous fusion method, with an increase in F-Score from 0.402 to 0.508. The significance of this establishes that the extraction of intermediate features produces a fertile environment for the detection of intermodal relationships for the task of LC classification. This research also provides an architecture to facilitate the future implementation of alternative biomarkers for lung cancer, one of the acknowledged limitations of this study.https://www.mdpi.com/2673-2688/3/3/38deep learninglung cancermachine learningmultimodal fusion |
| spellingShingle | James Barrett Thiago Viana EMM-LC Fusion: Enhanced Multimodal Fusion for Lung Cancer Classification AI deep learning lung cancer machine learning multimodal fusion |
| title | EMM-LC Fusion: Enhanced Multimodal Fusion for Lung Cancer Classification |
| title_full | EMM-LC Fusion: Enhanced Multimodal Fusion for Lung Cancer Classification |
| title_fullStr | EMM-LC Fusion: Enhanced Multimodal Fusion for Lung Cancer Classification |
| title_full_unstemmed | EMM-LC Fusion: Enhanced Multimodal Fusion for Lung Cancer Classification |
| title_short | EMM-LC Fusion: Enhanced Multimodal Fusion for Lung Cancer Classification |
| title_sort | emm lc fusion enhanced multimodal fusion for lung cancer classification |
| topic | deep learning lung cancer machine learning multimodal fusion |
| url | https://www.mdpi.com/2673-2688/3/3/38 |
| work_keys_str_mv | AT jamesbarrett emmlcfusionenhancedmultimodalfusionforlungcancerclassification AT thiagoviana emmlcfusionenhancedmultimodalfusionforlungcancerclassification |