Hyperspectral Imaging Reveals Spectral Differences and Can Distinguish Malignant Melanoma from Pigmented Basal Cell Carcinomas: A Pilot Study
Pigmented basal cell carcinomas can be difficult to distinguish from melanocytic tumours. Hyperspectral imaging is a non-invasive imaging technique that measures the reflectance spectra of skin in vivo. The aim of this prospective pilot study was to use a convolutional neural network classifier in h...
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Medical Journals Sweden
2021-02-01
|
| Series: | Acta Dermato-Venereologica |
| Subjects: | |
| Online Access: |
https://www.medicaljournals.se/acta/content/html/10.2340/00015555-3755
|
| _version_ | 1819279370046406656 |
|---|---|
| author | Janne Räsänen Mari Salmivuori Ilkka Pölönen Mari Grönroos Noora Neittaanmäki |
| author_facet | Janne Räsänen Mari Salmivuori Ilkka Pölönen Mari Grönroos Noora Neittaanmäki |
| author_sort | Janne Räsänen |
| collection | DOAJ |
| description | Pigmented basal cell carcinomas can be difficult to distinguish from melanocytic tumours. Hyperspectral imaging is a non-invasive imaging technique that measures the reflectance spectra of skin in vivo. The aim of this prospective pilot study was to use a convolutional neural network classifier in hyperspectral images for differential diagnosis between pigmented basal cell carcinomas and melanoma. A total of 26 pigmented lesions (10 pigmented basal cell carcinomas, 12 melanomas in situ, 4 invasive melanomas) were imaged with hyperspectral imaging and excised for histopathological diagnosis. For 2-class classifier (melanocytic tumours vs pigmented basal cell carcinomas) using the majority of the pixels to predict the class of the whole lesion, the results showed a sensitivity of 100% (95% confidence interval 81–100%), specificity of 90% (95% confidence interval 60–98%) and positive predictive value of 94% (95% confidence interval 73–99%). These results indicate that a convolutional neural network classifier can differentiate melanocytic tumours from pigmented basal cell carcinomas in hyperspectral images. Further studies are warranted in order to confirm these preliminary results, using larger samples and multiple tumour types, including all types of melanocytic lesions. |
| first_indexed | 2024-12-24T00:26:49Z |
| format | Article |
| id | doaj.art-0cce763b8bb34c2e93c50b87063f98ea |
| institution | Directory Open Access Journal |
| issn | 0001-5555 1651-2057 |
| language | English |
| last_indexed | 2024-12-24T00:26:49Z |
| publishDate | 2021-02-01 |
| publisher | Medical Journals Sweden |
| record_format | Article |
| series | Acta Dermato-Venereologica |
| spelling | doaj.art-0cce763b8bb34c2e93c50b87063f98ea2022-12-21T17:24:24ZengMedical Journals SwedenActa Dermato-Venereologica0001-55551651-20572021-02-011012adv0040510.2340/00015555-37556000Hyperspectral Imaging Reveals Spectral Differences and Can Distinguish Malignant Melanoma from Pigmented Basal Cell Carcinomas: A Pilot StudyJanne Räsänen0Mari SalmivuoriIlkka PölönenMari GrönroosNoora Neittaanmäki Department of Dermatology, Tampere University Hospital, FIN-33530 Tampere, Finland. E-mail: janne.rasanen@sll.fimnet.fi. Pigmented basal cell carcinomas can be difficult to distinguish from melanocytic tumours. Hyperspectral imaging is a non-invasive imaging technique that measures the reflectance spectra of skin in vivo. The aim of this prospective pilot study was to use a convolutional neural network classifier in hyperspectral images for differential diagnosis between pigmented basal cell carcinomas and melanoma. A total of 26 pigmented lesions (10 pigmented basal cell carcinomas, 12 melanomas in situ, 4 invasive melanomas) were imaged with hyperspectral imaging and excised for histopathological diagnosis. For 2-class classifier (melanocytic tumours vs pigmented basal cell carcinomas) using the majority of the pixels to predict the class of the whole lesion, the results showed a sensitivity of 100% (95% confidence interval 81–100%), specificity of 90% (95% confidence interval 60–98%) and positive predictive value of 94% (95% confidence interval 73–99%). These results indicate that a convolutional neural network classifier can differentiate melanocytic tumours from pigmented basal cell carcinomas in hyperspectral images. Further studies are warranted in order to confirm these preliminary results, using larger samples and multiple tumour types, including all types of melanocytic lesions. https://www.medicaljournals.se/acta/content/html/10.2340/00015555-3755 deep learning neural network basal cell carcinoma malignant melanoma |
| spellingShingle | Janne Räsänen Mari Salmivuori Ilkka Pölönen Mari Grönroos Noora Neittaanmäki Hyperspectral Imaging Reveals Spectral Differences and Can Distinguish Malignant Melanoma from Pigmented Basal Cell Carcinomas: A Pilot Study Acta Dermato-Venereologica deep learning neural network basal cell carcinoma malignant melanoma |
| title | Hyperspectral Imaging Reveals Spectral Differences and Can Distinguish Malignant Melanoma from Pigmented Basal Cell Carcinomas: A Pilot Study |
| title_full | Hyperspectral Imaging Reveals Spectral Differences and Can Distinguish Malignant Melanoma from Pigmented Basal Cell Carcinomas: A Pilot Study |
| title_fullStr | Hyperspectral Imaging Reveals Spectral Differences and Can Distinguish Malignant Melanoma from Pigmented Basal Cell Carcinomas: A Pilot Study |
| title_full_unstemmed | Hyperspectral Imaging Reveals Spectral Differences and Can Distinguish Malignant Melanoma from Pigmented Basal Cell Carcinomas: A Pilot Study |
| title_short | Hyperspectral Imaging Reveals Spectral Differences and Can Distinguish Malignant Melanoma from Pigmented Basal Cell Carcinomas: A Pilot Study |
| title_sort | hyperspectral imaging reveals spectral differences and can distinguish malignant melanoma from pigmented basal cell carcinomas a pilot study |
| topic | deep learning neural network basal cell carcinoma malignant melanoma |
| url |
https://www.medicaljournals.se/acta/content/html/10.2340/00015555-3755
|
| work_keys_str_mv | AT jannerasanen hyperspectralimagingrevealsspectraldifferencesandcandistinguishmalignantmelanomafrompigmentedbasalcellcarcinomasapilotstudy AT marisalmivuori hyperspectralimagingrevealsspectraldifferencesandcandistinguishmalignantmelanomafrompigmentedbasalcellcarcinomasapilotstudy AT ilkkapolonen hyperspectralimagingrevealsspectraldifferencesandcandistinguishmalignantmelanomafrompigmentedbasalcellcarcinomasapilotstudy AT marigronroos hyperspectralimagingrevealsspectraldifferencesandcandistinguishmalignantmelanomafrompigmentedbasalcellcarcinomasapilotstudy AT nooraneittaanmaki hyperspectralimagingrevealsspectraldifferencesandcandistinguishmalignantmelanomafrompigmentedbasalcellcarcinomasapilotstudy |