Effects of dimension reduction of hyperspectral images in skin gross pathology
© 2022 The Authors. Skin Research and Technology published by John Wiley & Sons Ltd..
BACKGROUND: Hyperspectral imaging (HSI) is an emerging modality for the gross pathology of the skin. Spectral signatures of HSI could discriminate malignant from benign tissue. Because of inherent redundancies in HSI and in order to facilitate the use of deep-learning models, dimension reduction is a common preprocessing step. The effects of dimension reduction choice, training scope, and number of retained dimensions have not been evaluated on skin HSI for segmentation tasks.
MATERIALS AND METHODS: An in-house dataset of HSI signatures from pigmented skin lesions was prepared and labeled with histology. Eleven different dimension reduction methods were used as preprocessing for tumor margin detection with support vector machines. Cluster-wise principal component analysis (ClusterPCA), a new variant of PCA, was proposed. The scope of application for dimension reduction was also investigated.
RESULTS: The components produced by ClusterPCA show good agreement with the expected optical properties of skin chromophores. Random forest importance performed best during classification. However, all methods suffered from low sensitivity and generalization.
CONCLUSION: Investigation of more complex reduction and segmentation schemes with emphasis on the nature of HSI and optical properties of the skin is necessary. Insights on dimension reduction for skin tissue could facilitate the development of HSI-based systems for cancer margin detection at gross level.
Medienart: |
E-Artikel |
---|
Erscheinungsjahr: |
2023 |
---|---|
Erschienen: |
2023 |
Enthalten in: |
Zur Gesamtaufnahme - volume:29 |
---|---|
Enthalten in: |
Skin research and technology : official journal of International Society for Bioengineering and the Skin (ISBS) [and] International Society for Digital Imaging of Skin (ISDIS) [and] International Society for Skin Imaging (ISSI) - 29(2023), 2 vom: 31. Feb., Seite e13270 |
Sprache: |
Englisch |
---|
Beteiligte Personen: |
Aloupogianni, Eleni [VerfasserIn] |
---|
Links: |
---|
Themen: |
Computer-assisted diagnosis |
---|
Anmerkungen: |
Date Completed 27.02.2023 Date Revised 12.08.2023 published: Print Citation Status MEDLINE |
---|
doi: |
10.1111/srt.13270 |
---|
funding: |
|
---|---|
Förderinstitution / Projekttitel: |
|
PPN (Katalog-ID): |
NLM353297399 |
---|
LEADER | 01000naa a22002652 4500 | ||
---|---|---|---|
001 | NLM353297399 | ||
003 | DE-627 | ||
005 | 20231226055653.0 | ||
007 | cr uuu---uuuuu | ||
008 | 231226s2023 xx |||||o 00| ||eng c | ||
024 | 7 | |a 10.1111/srt.13270 |2 doi | |
028 | 5 | 2 | |a pubmed24n1177.xml |
035 | |a (DE-627)NLM353297399 | ||
035 | |a (NLM)36823506 | ||
040 | |a DE-627 |b ger |c DE-627 |e rakwb | ||
041 | |a eng | ||
100 | 1 | |a Aloupogianni, Eleni |e verfasserin |4 aut | |
245 | 1 | 0 | |a Effects of dimension reduction of hyperspectral images in skin gross pathology |
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 27.02.2023 | ||
500 | |a Date Revised 12.08.2023 | ||
500 | |a published: Print | ||
500 | |a Citation Status MEDLINE | ||
520 | |a © 2022 The Authors. Skin Research and Technology published by John Wiley & Sons Ltd. | ||
520 | |a BACKGROUND: Hyperspectral imaging (HSI) is an emerging modality for the gross pathology of the skin. Spectral signatures of HSI could discriminate malignant from benign tissue. Because of inherent redundancies in HSI and in order to facilitate the use of deep-learning models, dimension reduction is a common preprocessing step. The effects of dimension reduction choice, training scope, and number of retained dimensions have not been evaluated on skin HSI for segmentation tasks | ||
520 | |a MATERIALS AND METHODS: An in-house dataset of HSI signatures from pigmented skin lesions was prepared and labeled with histology. Eleven different dimension reduction methods were used as preprocessing for tumor margin detection with support vector machines. Cluster-wise principal component analysis (ClusterPCA), a new variant of PCA, was proposed. The scope of application for dimension reduction was also investigated | ||
520 | |a RESULTS: The components produced by ClusterPCA show good agreement with the expected optical properties of skin chromophores. Random forest importance performed best during classification. However, all methods suffered from low sensitivity and generalization | ||
520 | |a CONCLUSION: Investigation of more complex reduction and segmentation schemes with emphasis on the nature of HSI and optical properties of the skin is necessary. Insights on dimension reduction for skin tissue could facilitate the development of HSI-based systems for cancer margin detection at gross level | ||
650 | 4 | |a Journal Article | |
650 | 4 | |a computer-assisted diagnosis | |
650 | 4 | |a dimension reduction | |
650 | 4 | |a gross pathology | |
650 | 4 | |a hyperspectral imaging | |
650 | 4 | |a segmentation | |
700 | 1 | |a Ishikawa, Masahiro |e verfasserin |4 aut | |
700 | 1 | |a Ichimura, Takaya |e verfasserin |4 aut | |
700 | 1 | |a Hamada, Mei |e verfasserin |4 aut | |
700 | 1 | |a Murakami, Takuo |e verfasserin |4 aut | |
700 | 1 | |a Sasaki, Atsushi |e verfasserin |4 aut | |
700 | 1 | |a Nakamura, Koichiro |e verfasserin |4 aut | |
700 | 1 | |a Kobayashi, Naoki |e verfasserin |4 aut | |
700 | 1 | |a Obi, Takashi |e verfasserin |4 aut | |
773 | 0 | 8 | |i Enthalten in |t Skin research and technology : official journal of International Society for Bioengineering and the Skin (ISBS) [and] International Society for Digital Imaging of Skin (ISDIS) [and] International Society for Skin Imaging (ISSI) |d 1995 |g 29(2023), 2 vom: 31. Feb., Seite e13270 |w (DE-627)NLM09446216X |x 1600-0846 |7 nnns |
773 | 1 | 8 | |g volume:29 |g year:2023 |g number:2 |g day:31 |g month:02 |g pages:e13270 |
856 | 4 | 0 | |u http://dx.doi.org/10.1111/srt.13270 |3 Volltext |
912 | |a GBV_USEFLAG_A | ||
912 | |a GBV_NLM | ||
951 | |a AR | ||
952 | |d 29 |j 2023 |e 2 |b 31 |c 02 |h e13270 |