Deep Learning to Measure the Intensity of Indocyanine Green in Endometriosis Surgeries with Intestinal Resection
Endometriosis is a gynecological pathology that affects between 6 and 15% of women of childbearing age. One of the manifestations is intestinal deep infiltrating endometriosis. This condition may force patients to resort to surgical treatment, often ending in resection. The level of blood perfusion...
| Main Authors: | , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2022-06-01
|
| Series: | Journal of Personalized Medicine |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2075-4426/12/6/982 |
| _version_ | 1797485499290484736 |
|---|---|
| author | Alicia Hernández Pablo Robles de Zulueta Emanuela Spagnolo Cristina Soguero Ignacio Cristobal Isabel Pascual Ana López David Ramiro-Cortijo |
| author_facet | Alicia Hernández Pablo Robles de Zulueta Emanuela Spagnolo Cristina Soguero Ignacio Cristobal Isabel Pascual Ana López David Ramiro-Cortijo |
| author_sort | Alicia Hernández |
| collection | DOAJ |
| description | Endometriosis is a gynecological pathology that affects between 6 and 15% of women of childbearing age. One of the manifestations is intestinal deep infiltrating endometriosis. This condition may force patients to resort to surgical treatment, often ending in resection. The level of blood perfusion at the anastomosis is crucial for its outcome, for this reason, indocyanine green (ICG), a fluorochrome that green stains the structures where it is present, is injected during surgery. This study proposes a novel method based on deep learning algorithms for quantifying the level of blood perfusion in anastomosis. Firstly, with a deep learning algorithm based on the U-Net, models capable of automatically segmenting the intestine from the surgical videos were generated. Secondly, blood perfusion level, from the already segmented video frames, was quantified. The frames were characterized using textures, precisely nine first- and second-order statistics, and then two experiments were carried out. In the first experiment, the differences in the perfusion between the two-anastomosis parts were determined, and in the second, it was verified that the ICG variation could be captured through the textures. The best model when segmenting has an accuracy of 0.92 and a dice coefficient of 0.96. It is concluded that segmentation of the bowel using the U-Net was successful, and the textures are appropriate descriptors for characterization of the blood perfusion in the images where ICG is present. This might help to predict whether postoperative complications will occur during surgery, enabling clinicians to act on this information. |
| first_indexed | 2024-03-09T23:19:41Z |
| format | Article |
| id | doaj.art-90bfdcb0516c4b6397c388b55f49399c |
| institution | Directory Open Access Journal |
| issn | 2075-4426 |
| language | English |
| last_indexed | 2024-03-09T23:19:41Z |
| publishDate | 2022-06-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Journal of Personalized Medicine |
| spelling | doaj.art-90bfdcb0516c4b6397c388b55f49399c2023-11-23T17:28:46ZengMDPI AGJournal of Personalized Medicine2075-44262022-06-0112698210.3390/jpm12060982Deep Learning to Measure the Intensity of Indocyanine Green in Endometriosis Surgeries with Intestinal ResectionAlicia Hernández0Pablo Robles de Zulueta1Emanuela Spagnolo2Cristina Soguero3Ignacio Cristobal4Isabel Pascual5Ana López6David Ramiro-Cortijo7Department of Obstetrics and Gynecology, Hospital Universitario La Paz, Paseo de la Castellana, 261, 28046 Madrid, SpainDepartment of Signal Theory and Communications, Telematics and Computing Systems, Universidad Rey Juan Carlos, Camino del Molino, 5, D201, Departamental III, 28942 Fuenlabrada, SpainDepartment of Obstetrics and Gynecology, Hospital Universitario La Paz, Paseo de la Castellana, 261, 28046 Madrid, SpainDepartment of Signal Theory and Communications, Telematics and Computing Systems, Universidad Rey Juan Carlos, Camino del Molino, 5, D201, Departamental III, 28942 Fuenlabrada, SpainDepartment of Obstetrics and Gynecology, Hospital Universitario La Paz, Paseo de la Castellana, 261, 28046 Madrid, SpainDepartment of General Surgery, Hospital Universitario La Paz, Paseo de la Castellana, 261, 28046 Madrid, SpainDepartment of Obstetrics and Gynecology, Hospital Universitario La Paz, Paseo de la Castellana, 261, 28046 Madrid, SpainDepartment of Physiology, Faculty of Medicine, Universidad Autónoma de Madrid, C/Arzobispo Morcillo 2, 28049 Madrid, SpainEndometriosis is a gynecological pathology that affects between 6 and 15% of women of childbearing age. One of the manifestations is intestinal deep infiltrating endometriosis. This condition may force patients to resort to surgical treatment, often ending in resection. The level of blood perfusion at the anastomosis is crucial for its outcome, for this reason, indocyanine green (ICG), a fluorochrome that green stains the structures where it is present, is injected during surgery. This study proposes a novel method based on deep learning algorithms for quantifying the level of blood perfusion in anastomosis. Firstly, with a deep learning algorithm based on the U-Net, models capable of automatically segmenting the intestine from the surgical videos were generated. Secondly, blood perfusion level, from the already segmented video frames, was quantified. The frames were characterized using textures, precisely nine first- and second-order statistics, and then two experiments were carried out. In the first experiment, the differences in the perfusion between the two-anastomosis parts were determined, and in the second, it was verified that the ICG variation could be captured through the textures. The best model when segmenting has an accuracy of 0.92 and a dice coefficient of 0.96. It is concluded that segmentation of the bowel using the U-Net was successful, and the textures are appropriate descriptors for characterization of the blood perfusion in the images where ICG is present. This might help to predict whether postoperative complications will occur during surgery, enabling clinicians to act on this information.https://www.mdpi.com/2075-4426/12/6/982deep endometriosisdeep learningvideo protocolautomatic segmentationbowel resectionlaparoscopy |
| spellingShingle | Alicia Hernández Pablo Robles de Zulueta Emanuela Spagnolo Cristina Soguero Ignacio Cristobal Isabel Pascual Ana López David Ramiro-Cortijo Deep Learning to Measure the Intensity of Indocyanine Green in Endometriosis Surgeries with Intestinal Resection Journal of Personalized Medicine deep endometriosis deep learning video protocol automatic segmentation bowel resection laparoscopy |
| title | Deep Learning to Measure the Intensity of Indocyanine Green in Endometriosis Surgeries with Intestinal Resection |
| title_full | Deep Learning to Measure the Intensity of Indocyanine Green in Endometriosis Surgeries with Intestinal Resection |
| title_fullStr | Deep Learning to Measure the Intensity of Indocyanine Green in Endometriosis Surgeries with Intestinal Resection |
| title_full_unstemmed | Deep Learning to Measure the Intensity of Indocyanine Green in Endometriosis Surgeries with Intestinal Resection |
| title_short | Deep Learning to Measure the Intensity of Indocyanine Green in Endometriosis Surgeries with Intestinal Resection |
| title_sort | deep learning to measure the intensity of indocyanine green in endometriosis surgeries with intestinal resection |
| topic | deep endometriosis deep learning video protocol automatic segmentation bowel resection laparoscopy |
| url | https://www.mdpi.com/2075-4426/12/6/982 |
| work_keys_str_mv | AT aliciahernandez deeplearningtomeasuretheintensityofindocyaninegreeninendometriosissurgerieswithintestinalresection AT pabloroblesdezulueta deeplearningtomeasuretheintensityofindocyaninegreeninendometriosissurgerieswithintestinalresection AT emanuelaspagnolo deeplearningtomeasuretheintensityofindocyaninegreeninendometriosissurgerieswithintestinalresection AT cristinasoguero deeplearningtomeasuretheintensityofindocyaninegreeninendometriosissurgerieswithintestinalresection AT ignaciocristobal deeplearningtomeasuretheintensityofindocyaninegreeninendometriosissurgerieswithintestinalresection AT isabelpascual deeplearningtomeasuretheintensityofindocyaninegreeninendometriosissurgerieswithintestinalresection AT analopez deeplearningtomeasuretheintensityofindocyaninegreeninendometriosissurgerieswithintestinalresection AT davidramirocortijo deeplearningtomeasuretheintensityofindocyaninegreeninendometriosissurgerieswithintestinalresection |