1
|
Fang YJ, Huang CW, Karmakar R, Mukundan A, Tsao YM, Yang KY, Wang HC. Assessment of Narrow-Band Imaging Algorithm for Video Capsule Endoscopy Based on Decorrelated Color Space for Esophageal Cancer: Part II, Detection and Classification of Esophageal Cancer. Cancers (Basel) 2024; 16:572. [PMID: 38339322 PMCID: PMC10854620 DOI: 10.3390/cancers16030572] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Grants] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 01/07/2024] [Revised: 01/26/2024] [Accepted: 01/26/2024] [Indexed: 02/12/2024] Open
Abstract
Esophageal carcinoma (EC) is a prominent contributor to cancer-related mortality since it lacks discernible features in its first phases. Multiple studies have shown that narrow-band imaging (NBI) has superior accuracy, sensitivity, and specificity in detecting EC compared to white light imaging (WLI). Thus, this study innovatively employs a color space linked to décor to transform WLIs into NBIs, offering a novel approach to enhance the detection capabilities of EC in its early stages. In this study a total of 3415 WLI along with the corresponding 3415 simulated NBI images were used for analysis combined with the YOLOv5 algorithm to train the WLI images and the NBI images individually showcasing the adaptability of advanced object detection techniques in the context of medical image analysis. The evaluation of the model's performance was based on the produced confusion matrix and five key metrics: precision, recall, specificity, accuracy, and F1-score of the trained model. The model underwent training to accurately identify three specific manifestations of EC, namely dysplasia, squamous cell carcinoma (SCC), and polyps demonstrates a nuanced and targeted analysis, addressing diverse aspects of EC pathology for a more comprehensive understanding. The NBI model effectively enhanced both its recall and accuracy rates in detecting dysplasia cancer, a pre-cancerous stage that might improve the overall five-year survival rate. Conversely, the SCC category decreased its accuracy and recall rate, although the NBI and WLI models performed similarly in recognizing the polyp. The NBI model demonstrated an accuracy of 0.60, 0.81, and 0.66 in the dysplasia, SCC, and polyp categories, respectively. Additionally, it attained a recall rate of 0.40, 0.73, and 0.76 in the same categories. The WLI model demonstrated an accuracy of 0.56, 0.99, and 0.65 in the dysplasia, SCC, and polyp categories, respectively. Additionally, it obtained a recall rate of 0.39, 0.86, and 0.78 in the same categories, respectively. The limited number of training photos is the reason for the suboptimal performance of the NBI model which can be improved by increasing the dataset.
Collapse
Affiliation(s)
- Yu-Jen Fang
- Department of Internal Medicine, National Taiwan University Hospital, Yun-Lin Branch, No. 579, Sec. 2, Yunlin Rd., Dou-Liu 64041, Taiwan;
- Department of Internal Medicine, National Taiwan University College of Medicine, No. 1, Jen Ai Rd., Sec. 1, Taipei 10051, Taiwan
| | - Chien-Wei Huang
- Department of Gastroenterology, Kaohsiung Armed Forces General Hospital, 2, Zhongzheng 1st Rd., Lingya District, Kaohsiung 80284, Taiwan;
- Department of Nursing, Tajen University, 20, Weixin Rd., Yanpu Township, Pingtung County 90741, Taiwan
| | - Riya Karmakar
- Department of Mechanical Engineering, National Chung Cheng University, 168, University Rd., Min Hsiung, Chia Yi 62102, Taiwan; (R.K.); (A.M.); (Y.-M.T.)
| | - Arvind Mukundan
- Department of Mechanical Engineering, National Chung Cheng University, 168, University Rd., Min Hsiung, Chia Yi 62102, Taiwan; (R.K.); (A.M.); (Y.-M.T.)
| | - Yu-Ming Tsao
- Department of Mechanical Engineering, National Chung Cheng University, 168, University Rd., Min Hsiung, Chia Yi 62102, Taiwan; (R.K.); (A.M.); (Y.-M.T.)
| | - Kai-Yao Yang
- Department of Gastroenterology, Kaohsiung Armed Forces General Hospital, 2, Zhongzheng 1st Rd., Lingya District, Kaohsiung 80284, Taiwan;
| | - Hsiang-Chen Wang
- Department of Mechanical Engineering, National Chung Cheng University, 168, University Rd., Min Hsiung, Chia Yi 62102, Taiwan; (R.K.); (A.M.); (Y.-M.T.)
- Department of Medical Research, Dalin Tzu Chi Hospital, Buddhist Tzu Chi Medical Foundation, No. 2, Minsheng Road, Dalin, Chia Yi 62247, Taiwan
- Hitspectra Intelligent Technology Co., Ltd., 4F, No. 2, Fuxing 4th Rd., Qianzhen District, Kaohsiung 80661, Taiwan
| |
Collapse
|
3
|
Qiu L, Chuttani R, Pleskow DK, Turzhitsky V, Khan U, Zakharov YN, Zhang L, Berzin TM, Yee EU, Sawhney MS, Li Y, Vitkin E, Goldsmith JD, Itzkan I, Perelman LT. Multispectral light scattering endoscopic imaging of esophageal precancer. LIGHT, SCIENCE & APPLICATIONS 2018; 7:17174. [PMID: 30839534 PMCID: PMC6060057 DOI: 10.1038/lsa.2017.174] [Citation(s) in RCA: 12] [Impact Index Per Article: 2.0] [Reference Citation Analysis] [Abstract] [Key Words] [Grants] [Track Full Text] [Subscribe] [Scholar Register] [Received: 07/13/2017] [Revised: 12/18/2017] [Accepted: 12/18/2017] [Indexed: 05/14/2023]
Abstract
Esophageal adenocarcinoma is the most rapidly growing cancer in America. Although the prognosis after diagnosis is unfavorable, the chance of a successful outcome increases tremendously if detected early while the lesion is still dysplastic. Unfortunately, the present standard-of-care, endoscopic surveillance, has major limitations, since dysplasia is invisible, often focal, and systematic biopsies typically sample less than one percent of the esophageal lining and therefore easily miss malignancies. To solve this problem we developed a multispectral light scattering endoscopic imaging system. It surveys the entire esophageal lining and accurately detects subcellular dysplastic changes. The system combines light scattering spectroscopy, which detects and identifies invisible dysplastic sites by analyzing light scattered from epithelial cells, with rapid scanning of the entire esophageal lining using a collimated broadband light beam delivered by an endoscopically compatible fiber optic probe. Here we report the results of the first comprehensive multispectral imaging study, conducted as part of routine endoscopic procedures performed on patients with suspected dysplasia. In a double-blind study that characterized the system's ability to serve as a screening tool, 55 out of 57 patients were diagnosed correctly. In addition, a smaller double-blind comparison of the multispectral data in 24 patients with subsequent pathology at locations where 411 biopsies were collected yielded an accuracy of 90% in detecting individual locations of dysplasia, demonstrating the capability of this method to serve as a guide for biopsy.
Collapse
Affiliation(s)
- Le Qiu
- Center for Advanced Biomedical Imaging and Photonics, Division of Gastroenterology, Department of Medicine, Beth Israel Deaconess Medical Center, Harvard University, Boston, MA 02215, USA
| | - Ram Chuttani
- Division of Gastroenterology, Department of Medicine, Beth Israel Deaconess Medical Center, Harvard University, Boston, MA 02215, USA
| | - Douglas K Pleskow
- Division of Gastroenterology, Department of Medicine, Beth Israel Deaconess Medical Center, Harvard University, Boston, MA 02215, USA
| | - Vladimir Turzhitsky
- Center for Advanced Biomedical Imaging and Photonics, Division of Gastroenterology, Department of Medicine, Beth Israel Deaconess Medical Center, Harvard University, Boston, MA 02215, USA
| | - Umar Khan
- Center for Advanced Biomedical Imaging and Photonics, Division of Gastroenterology, Department of Medicine, Beth Israel Deaconess Medical Center, Harvard University, Boston, MA 02215, USA
| | - Yuri N Zakharov
- Center for Advanced Biomedical Imaging and Photonics, Division of Gastroenterology, Department of Medicine, Beth Israel Deaconess Medical Center, Harvard University, Boston, MA 02215, USA
| | - Lei Zhang
- Center for Advanced Biomedical Imaging and Photonics, Division of Gastroenterology, Department of Medicine, Beth Israel Deaconess Medical Center, Harvard University, Boston, MA 02215, USA
| | - Tyler M Berzin
- Division of Gastroenterology, Department of Medicine, Beth Israel Deaconess Medical Center, Harvard University, Boston, MA 02215, USA
| | - Eric U Yee
- Department of Pathology, Beth Israel Deaconess Medical Center, Harvard University, Boston, MA 02215, USA
| | - Mandeep S Sawhney
- Division of Gastroenterology, Department of Medicine, Beth Israel Deaconess Medical Center, Harvard University, Boston, MA 02215, USA
| | - Yunping Li
- Department of Anesthesia, Critical Care and Pain Medicine, Beth Israel Deaconess Medical Center, Harvard University, Boston, MA 02215, USA
| | - Edward Vitkin
- Center for Advanced Biomedical Imaging and Photonics, Division of Gastroenterology, Department of Medicine, Beth Israel Deaconess Medical Center, Harvard University, Boston, MA 02215, USA
| | - Jeffrey D Goldsmith
- Department of Pathology, Beth Israel Deaconess Medical Center, Harvard University, Boston, MA 02215, USA
| | - Irving Itzkan
- Center for Advanced Biomedical Imaging and Photonics, Division of Gastroenterology, Department of Medicine, Beth Israel Deaconess Medical Center, Harvard University, Boston, MA 02215, USA
| | - Lev T Perelman
- Center for Advanced Biomedical Imaging and Photonics, Division of Gastroenterology, Department of Medicine, Beth Israel Deaconess Medical Center, Harvard University, Boston, MA 02215, USA
- Division of Gastroenterology, Department of Medicine, Beth Israel Deaconess Medical Center, Harvard University, Boston, MA 02215, USA
- Biological and Biomedical Sciences Program, Harvard University, Boston, MA 02215, USA
| |
Collapse
|