Multimodal integration of image, epigenetic and clinical data to predict BRAF mutation status in melanoma
Copyright © 2023 The Authors. Published by Elsevier Ltd.. All rights reserved..
BACKGROUND: In machine learning, multimodal classifiers can provide more generalised performance than unimodal classifiers. In clinical practice, physicians usually also rely on a range of information from different examinations for diagnosis. In this study, we used BRAF mutation status prediction in melanoma as a model system to analyse the contribution of different data types in a combined classifier because BRAF status can be determined accurately by sequencing as the current gold standard, thus nearly eliminating label noise.
METHODS: We trained a deep learning-based classifier by combining individually trained random forests of image, clinical and methylation data to predict BRAF-V600 mutation status in primary and metastatic melanomas of The Cancer Genome Atlas cohort.
RESULTS: With our multimodal approach, we achieved an area under the receiver operating characteristic curve of 0.80, whereas the individual classifiers yielded areas under the receiver operating characteristic curve of 0.63 (histopathologic image data), 0.66 (clinical data) and 0.66 (methylation data) on an independent data set.
CONCLUSIONS: Our combined approach can predict BRAF status to some extent by identifying BRAF-V600 specific patterns at the histologic, clinical and epigenetic levels. The multimodal classifiers have improved generalisability in predicting BRAF mutation status.
| Medienart: |
E-Artikel |
|---|
| Erscheinungsjahr: |
2023 |
|---|---|
| Erschienen: |
2023 |
| Enthalten in: |
Zur Gesamtaufnahme - volume:183 |
|---|---|
| Enthalten in: |
European journal of cancer (Oxford, England : 1990) - 183(2023) vom: 30. Apr., Seite 131-138 |
| Sprache: |
Englisch |
|---|
| Beteiligte Personen: |
Schneider, Lucas [VerfasserIn] |
|---|
| Links: |
|---|
| Themen: |
BRAF protein, human |
|---|
| Anmerkungen: |
Date Completed 20.03.2023 Date Revised 25.03.2023 published: Print-Electronic Citation Status MEDLINE |
|---|
| doi: |
10.1016/j.ejca.2023.01.021 |
|---|
| funding: |
|
|---|---|
| Förderinstitution / Projekttitel: |
|
| PPN (Katalog-ID): |
NLM353602655 |
|---|
| LEADER | 01000naa a22002652 4500 | ||
|---|---|---|---|
| 001 | NLM353602655 | ||
| 003 | DE-627 | ||
| 005 | 20231226060402.0 | ||
| 007 | cr uuu---uuuuu | ||
| 008 | 231226s2023 xx |||||o 00| ||eng c | ||
| 024 | 7 | |a 10.1016/j.ejca.2023.01.021 |2 doi | |
| 028 | 5 | 2 | |a pubmed24n1178.xml |
| 035 | |a (DE-627)NLM353602655 | ||
| 035 | |a (NLM)36854237 | ||
| 035 | |a (PII)S0959-8049(23)00045-X | ||
| 040 | |a DE-627 |b ger |c DE-627 |e rakwb | ||
| 041 | |a eng | ||
| 100 | 1 | |a Schneider, Lucas |e verfasserin |4 aut | |
| 245 | 1 | 0 | |a Multimodal integration of image, epigenetic and clinical data to predict BRAF mutation status in melanoma |
| 264 | 1 | |c 2023 | |
| 336 | |a Text |b txt |2 rdacontent | ||
| 337 | |a ƒaComputermedien |b c |2 rdamedia | ||
| 338 | |a ƒa Online-Ressource |b cr |2 rdacarrier | ||
| 500 | |a Date Completed 20.03.2023 | ||
| 500 | |a Date Revised 25.03.2023 | ||
| 500 | |a published: Print-Electronic | ||
| 500 | |a Citation Status MEDLINE | ||
| 520 | |a Copyright © 2023 The Authors. Published by Elsevier Ltd.. All rights reserved. | ||
| 520 | |a BACKGROUND: In machine learning, multimodal classifiers can provide more generalised performance than unimodal classifiers. In clinical practice, physicians usually also rely on a range of information from different examinations for diagnosis. In this study, we used BRAF mutation status prediction in melanoma as a model system to analyse the contribution of different data types in a combined classifier because BRAF status can be determined accurately by sequencing as the current gold standard, thus nearly eliminating label noise | ||
| 520 | |a METHODS: We trained a deep learning-based classifier by combining individually trained random forests of image, clinical and methylation data to predict BRAF-V600 mutation status in primary and metastatic melanomas of The Cancer Genome Atlas cohort | ||
| 520 | |a RESULTS: With our multimodal approach, we achieved an area under the receiver operating characteristic curve of 0.80, whereas the individual classifiers yielded areas under the receiver operating characteristic curve of 0.63 (histopathologic image data), 0.66 (clinical data) and 0.66 (methylation data) on an independent data set | ||
| 520 | |a CONCLUSIONS: Our combined approach can predict BRAF status to some extent by identifying BRAF-V600 specific patterns at the histologic, clinical and epigenetic levels. The multimodal classifiers have improved generalisability in predicting BRAF mutation status | ||
| 650 | 4 | |a Journal Article | |
| 650 | 4 | |a Research Support, Non-U.S. Gov't | |
| 650 | 4 | |a BRAF-V600 mutation | |
| 650 | 4 | |a Deep learning | |
| 650 | 4 | |a Melanoma | |
| 650 | 4 | |a Multimodal classifier | |
| 650 | 4 | |a Mutation prediction | |
| 650 | 7 | |a Proto-Oncogene Proteins B-raf |2 NLM | |
| 650 | 7 | |a EC 2.7.11.1 |2 NLM | |
| 650 | 7 | |a BRAF protein, human |2 NLM | |
| 650 | 7 | |a EC 2.7.11.1 |2 NLM | |
| 700 | 1 | |a Wies, Christoph |e verfasserin |4 aut | |
| 700 | 1 | |a Krieghoff-Henning, Eva I |e verfasserin |4 aut | |
| 700 | 1 | |a Bucher, Tabea-Clara |e verfasserin |4 aut | |
| 700 | 1 | |a Utikal, Jochen S |e verfasserin |4 aut | |
| 700 | 1 | |a Schadendorf, Dirk |e verfasserin |4 aut | |
| 700 | 1 | |a Brinker, Titus J |e verfasserin |4 aut | |
| 773 | 0 | 8 | |i Enthalten in |t European journal of cancer (Oxford, England : 1990) |d 1991 |g 183(2023) vom: 30. Apr., Seite 131-138 |w (DE-627)NLM012602779 |x 1879-0852 |7 nnns |
| 773 | 1 | 8 | |g volume:183 |g year:2023 |g day:30 |g month:04 |g pages:131-138 |
| 856 | 4 | 0 | |u http://dx.doi.org/10.1016/j.ejca.2023.01.021 |3 Volltext |
| 912 | |a GBV_USEFLAG_A | ||
| 912 | |a GBV_NLM | ||
| 951 | |a AR | ||
| 952 | |d 183 |j 2023 |b 30 |c 04 |h 131-138 | ||