1
|
Deep learning diagnostics for bladder tumor identification and grade prediction using RGB method. Sci Rep 2022; 12:17699. [PMID: 36271252 PMCID: PMC9587038 DOI: 10.1038/s41598-022-22797-7] [Citation(s) in RCA: 9] [Impact Index Per Article: 4.5] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 05/09/2022] [Accepted: 10/19/2022] [Indexed: 01/18/2023] Open
Abstract
We evaluate the diagnostic performance of deep learning artificial intelligence (AI) for bladder cancer, which used white-light images (WLIs) and narrow-band images, and tumor grade prediction of AI based on tumor color using the red/green/blue (RGB) method. This retrospective study analyzed 10,991 cystoscopic images of suspicious bladder tumors using a mask region-based convolutional neural network with a ResNeXt-101-32 × 8d-FPN backbone. The diagnostic performance of AI was evaluated by calculating sensitivity, specificity, and diagnostic accuracy, and its ability to detect cancers was investigated using the dice score coefficient (DSC). Using the support vector machine model, we analyzed differences in tumor colors according to tumor grade using the RGB method. The sensitivity, specificity, diagnostic accuracy and DSC of AI were 95.0%, 93.7%, 94.1% and 74.7%. In WLIs, there were differences in red and blue values according to tumor grade (p < 0.001). According to the average RGB value, the performance was ≥ 98% for the diagnosis of benign vs. low-and high-grade tumors using WLIs and > 90% for the diagnosis of chronic non-specific inflammation vs. carcinoma in situ using WLIs. The diagnostic performance of the AI-assisted diagnosis was of high quality, and the AI could distinguish the tumor grade based on tumor color.
Collapse
|
2
|
Kalinichev AV, Kravchenko AV, Gryazev IP, Kechin AA, Karpukhin OR, Khairullina EM, Kartsova LA, Golovkina AG, Kozynchenko VA, Peshkova MA, Tumkin II. Classification of ballpoint pen inks based on selective extraction and subsequent digital color and cluster analyses. Analyst 2022; 147:3055-3064. [DOI: 10.1039/d2an00482h] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 11/21/2022]
Abstract
Extraction and color analysis coupled with machine learning allows clustering of pen inks and realizing preliminary classification when assessing document age.
Collapse
Affiliation(s)
- Andrey V. Kalinichev
- Saint Petersburg State University, 7/9 Universitetskaya nab., St. Petersburg 199034, Russia
| | | | - Ivan P. Gryazev
- Saint Petersburg State University, 7/9 Universitetskaya nab., St. Petersburg 199034, Russia
| | - Arseniy A. Kechin
- Saint Petersburg State University, 7/9 Universitetskaya nab., St. Petersburg 199034, Russia
| | - Oleg R. Karpukhin
- Saint Petersburg State University, 7/9 Universitetskaya nab., St. Petersburg 199034, Russia
| | | | - Liudmila A. Kartsova
- Saint Petersburg State University, 7/9 Universitetskaya nab., St. Petersburg 199034, Russia
| | - Anna G. Golovkina
- Saint Petersburg State University, 7/9 Universitetskaya nab., St. Petersburg 199034, Russia
| | | | - Maria A. Peshkova
- Saint Petersburg State University, 7/9 Universitetskaya nab., St. Petersburg 199034, Russia
| | - Ilya I. Tumkin
- Saint Petersburg State University, 7/9 Universitetskaya nab., St. Petersburg 199034, Russia
| |
Collapse
|
3
|
Thompson NBA, O'Sullivan SE, Howell RJ, Bailey DJ, Gilbert MR, Hyatt NC. Objective colour analysis from digital images as a nuclear forensic tool. Forensic Sci Int 2020; 319:110678. [PMID: 33444895 DOI: 10.1016/j.forsciint.2020.110678] [Citation(s) in RCA: 5] [Impact Index Per Article: 1.3] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 07/05/2020] [Revised: 12/22/2020] [Accepted: 12/23/2020] [Indexed: 02/05/2023]
Abstract
A digital colour image may be composed of hundreds of thousands of pixels, every pixel exhibiting a single colour. Each colour can be described as a combination of red, green and blue (RGB) components, of discrete values between 0-255. The RGB data contained within the pixels of an image could, therefore, be used to quantitatively establish the colour of nuclear material powders from digital images, particularly for use in nuclear forensics applications, where there is a need for consistent, objective analysis. This paper sets out a standard method for the photography and analysis of digital images of uranium oxide powder, for the objective quantification of colour by mean RGB values. Eight heat treated (up to 550°C) powder samples of studtite ([(UO2)(O2)(H2O)2]·2H2O) were photographed at room temperature and analysed by the RGB method. Hue, saturation and value of the coloured samples were obtained alongside mean RGB values, both of which were used to successfully determine the heating temperatures of unknown specimens of studtite.
Collapse
Affiliation(s)
- Nathan B A Thompson
- University of Sheffield, Department of Materials Science and Engineering, Sir Robert Hadfield Building, Sheffield, S1 3JD, UK.
| | - Sarah E O'Sullivan
- University of Sheffield, Department of Materials Science and Engineering, Sir Robert Hadfield Building, Sheffield, S1 3JD, UK.
| | - Robert J Howell
- University of Sheffield, Department of Materials Science and Engineering, Sir Robert Hadfield Building, Sheffield, S1 3JD, UK.
| | - Daniel J Bailey
- University of Sheffield, Department of Materials Science and Engineering, Sir Robert Hadfield Building, Sheffield, S1 3JD, UK.
| | | | - Neil C Hyatt
- University of Sheffield, Department of Materials Science and Engineering, Sir Robert Hadfield Building, Sheffield, S1 3JD, UK.
| |
Collapse
|
4
|
Zheng J, Huo D, Wen H, Shang Q, Sun W, Xu Z. Corneal-Smart Phone: A novel method to intelligently estimate postmortem interval. J Forensic Sci 2020; 66:356-364. [PMID: 33112427 DOI: 10.1111/1556-4029.14611] [Citation(s) in RCA: 3] [Impact Index Per Article: 0.8] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 07/30/2020] [Revised: 10/07/2020] [Accepted: 10/08/2020] [Indexed: 11/29/2022]
Abstract
The changes of postmortem corneal opacity are often used to roughly estimate the postmortem interval (PMI) in forensic practice. The difficulty associated with this time estimate is the lack of objective means to rapidly quantify postmortem corneal changes in crime scenes. This study constructed a data analysis model of PMI estimation and implemented an intelligent analysis system for examining the sequential changes of postmortem corneal digital images, named Corneal-Smart Phone, which can be used to quickly estimate PMI. The smart phone was used in combination with an attachment device that provided a darkroom environment and a steady light source to capture postmortem corneal images. By segmenting the corneal pupil region images, six color features, Red (R), Green (G), Blue (B), Hue (H), Saturation (S), Brightness (V) and four texture features Contrast (CON), Correlation (COR), Angular Second Moment (ASM), and Homogeneity (HOM), were extracted and correlated with PMI model. The results indicated that CON had the highest correlation with PMI (R2 = 0.983). No intra/intersubject variation in CON values were observed (p > 0.05). With the increase in ambient temperature or the decrease in humidity, the CON values were increased. PMI prediction error was <3 h within 36 h postmortem and extended to about 6-8 h after 36 h postmortem. The correct classification rate of the blind test samples was 82%. Our study provides a method that combines postmortem corneal image acquisition and digital image analysis to enable users to quickly obtain PMI estimation.
Collapse
Affiliation(s)
- JiLong Zheng
- Department of Forensic Medicine, Criminal Investigation Police University of China, Shenyang, China
| | - DeMin Huo
- Criminal Investigation Division of Jiading District Bureau of Shanghai Public Security Bureau, Shanghai, China
| | - HongYang Wen
- Center of Software Research and Development, Criminal Investigation Police University of China, Shenyang, China
| | - QingFa Shang
- Department of Forensic Medicine, Criminal Investigation Police University of China, Shenyang, China
| | - WenKai Sun
- Department of Forensic Medicine, Criminal Investigation Police University of China, Shenyang, China
| | - ZiTong Xu
- Department of Forensic Medicine, Criminal Investigation Police University of China, Shenyang, China
| |
Collapse
|
5
|
Tezuka T, Inayama M, Suzue R, Miyamoto K, Haku T. A Tuberculous Bronchial Artery Aneurysm with Abnormal Findings on Autofluorescence Imaging Bronchoscopy. Intern Med 2020; 59:1629-1632. [PMID: 32238720 PMCID: PMC7402969 DOI: 10.2169/internalmedicine.3610-19] [Citation(s) in RCA: 2] [Impact Index Per Article: 0.5] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Submit a Manuscript] [Subscribe] [Scholar Register] [Indexed: 01/30/2023] Open
Abstract
Pulmonary tuberculosis is a common disease that may result in hemoptysis. Fetal hemoptysis is known to be related to the rupture of a pulmonary aneurysm formed in the cavity wall. We herein report a case of non-cavity pulmonary tuberculosis that developed with massive hemoptysis following bronchial artery aneurysm. Bronchial artery embolization was performed, and autofluorescence imaging bronchoscopy was conducted one month after the anti-tuberculosis treatment. Bright-green color was observed in the ulcerative lesion with a white coat, corresponding to the bronchial artery aneurysm. This is the first report of the autofluorescence imaging observation of an ulcerative lesion caused by bronchial tuberculosis.
Collapse
Affiliation(s)
- Toshifumi Tezuka
- Department of Respiratory Medicine, Tokushima Prefectural Central Hospital, Japan
| | - Mami Inayama
- Department of Respiratory Medicine, Tokushima Prefectural Central Hospital, Japan
| | - Ryoko Suzue
- Department of Respiratory Medicine, Tokushima Prefectural Central Hospital, Japan
| | - Kenya Miyamoto
- Department of Respiratory Medicine, Tokushima Prefectural Central Hospital, Japan
| | - Takashi Haku
- Department of Respiratory Medicine, Tokushima Prefectural Central Hospital, Japan
| |
Collapse
|
6
|
Retraction statement: RGB and HSV quantitative analysis of autofluorescence bronchoscopy used for characterization and identification of bronchopulmonary cancer. Cancer Med 2020; 9:3634. [PMID: 32406178 PMCID: PMC7221306 DOI: 10.1002/cam4.2961] [Citation(s) in RCA: 1] [Impact Index Per Article: 0.3] [Reference Citation Analysis] [Track Full Text] [Download PDF] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 01/29/2023] Open
|
7
|
Saito N, Yamasaki M, Daido W, Ishiyama S, Deguchi N, Taniwaki M. A bronchial fibroepithelial polyp with abnormal findings on auto-fluorescence imaging. Respirol Case Rep 2017; 5:e00244. [PMID: 28603622 PMCID: PMC5465754 DOI: 10.1002/rcr2.244] [Citation(s) in RCA: 3] [Impact Index Per Article: 0.4] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 01/28/2017] [Revised: 04/04/2017] [Accepted: 05/04/2017] [Indexed: 01/31/2023] Open
Abstract
Bronchial fibroepithelial polyps represent a rare type of tumour that displays endobronchial growth. The findings of these lesions on auto‐fluorescence imaging (AFI) bronchoscopy have not been reported, despite the usefulness of AFI in detecting early lung cancer. We report the case of a patient with a bronchial fibroepithelial polyp that displayed positivity (magenta colour) on AFI. The patient was a 65‐year‐old man, in whom an endobronchial polypoid lesion of 10 mm diameter had been detected in the right basal bronchus by chest computed tomography (CT). On bronchoscopic examination, we found a whitish, smooth polypoid lesion. The lesion appeared magenta on AFI. On CT, however, the lesion had been almost stable for 4 years and 4 months. Bronchial fibroepithelial polyps may show AFI positivity, even when the lesion displays benign behaviour. The diagnosis of the lesion should not be confused by AFI positivity, and unnecessary surgical intervention should be avoided.
Collapse
Affiliation(s)
- Naomi Saito
- Department of Respiratory DiseaseHiroshima Red Cross Hospital & Atomic Bomb Survivors HospitalHiroshimaJapan
| | - Masahiro Yamasaki
- Department of Respiratory DiseaseHiroshima Red Cross Hospital & Atomic Bomb Survivors HospitalHiroshimaJapan
| | - Wakako Daido
- Department of Respiratory DiseaseHiroshima Red Cross Hospital & Atomic Bomb Survivors HospitalHiroshimaJapan
| | - Sayaka Ishiyama
- Department of Respiratory DiseaseHiroshima Red Cross Hospital & Atomic Bomb Survivors HospitalHiroshimaJapan
| | - Naoko Deguchi
- Department of Respiratory DiseaseHiroshima Red Cross Hospital & Atomic Bomb Survivors HospitalHiroshimaJapan
| | - Masaya Taniwaki
- Department of Respiratory DiseaseHiroshima Red Cross Hospital & Atomic Bomb Survivors HospitalHiroshimaJapan
| |
Collapse
|
8
|
Zheng X, Xiong H, Li Y, Han B, Sun J. RGB and HSV quantitative analysis of autofluorescence bronchoscopy used for characterization and identification of bronchopulmonary cancer. Cancer Med 2016; 5:3023-3030. [PMID: 27709786 PMCID: PMC5119956 DOI: 10.1002/cam4.831] [Citation(s) in RCA: 11] [Impact Index Per Article: 1.4] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 01/06/2016] [Revised: 06/02/2016] [Accepted: 06/13/2016] [Indexed: 01/11/2023] Open
Abstract
Autofluorescence bronchoscopy (AFB) shows good sensitivity in detecting dysplasia and bronchopulmonary cancer. However, the poor specificity of AFB would lead to excessive biopsy. The aim of the study is to establish a more effective quantitative method (optimal identification index and reference value) for characterizing the AFB images within the region of interest and discuss AFB's significance in the diagnosis of central‐type lung cancer. A total of 218 suspected lung cancer patients were enrolled in this study. A quantitative analysis based on color space (red, green, blue[RGB] and HSV system) was conducted and the result was compared with the final diagnosis obtained by the pathology of biopsy. Cases were divided into different groups according to the pathological diagnosis of normal bronchial mucosa, inflammation, low‐grade preinvasive (LGD), high‐grade preinvasive (HGD), and invasive cancer. Quantitative analyses in multi‐color spaces for the lesions showed by AFB images were conducted by software MATLAB. Finally, there is statistical significance among the different groups in some parameter in RGB and HSV system. So, both RGB and HSV quantitative analysis of autofluorescence bronchoscopy are useful to define benign and malignant diseases, which can objectively guide the bronchoscopist in selecting sites for biopsy with good pathologic correlation.
Collapse
Affiliation(s)
- Xiaoxuan Zheng
- Department of Endoscopy, Shanghai Chest Hospital, Shanghai Jiao Tong University, Shanghai 200030, China
| | - Hongkai Xiong
- Department of Electronic Engineering, Shanghai Jiao Tong University, Shanghai 200240, China
| | - Yong Li
- Department of Endoscopy, Shanghai Chest Hospital, Shanghai Jiao Tong University, Shanghai 200030, China
| | - Baohui Han
- Department of pulmonary medicine, Shanghai Chest Hospital, Shanghai Jiao Tong University, Shanghai 200030, China
| | - Jiayuan Sun
- Department of Endoscopy, Shanghai Chest Hospital, Shanghai Jiao Tong University, Shanghai 200030, China
| |
Collapse
|