Deep learning for gastroscopic images: computer-aided techniques for clinicians

Jin, Ziyi; Gan, Tianyuan; Wang, Peng; Fu, Zuoming; Zhang, Chongan; Yan, Qinglai; Zheng, Xue-Yong; Liang, Xiao; Ye, Xuesong

doi:10.1186/s12938-022-00979-8

Cited by 20 publications

(4 citation statements)

References 112 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In recent years, artificial intelligence (AI) has been making remarkable progress in various fields, including healthcare 7 . Researchers are increasingly using AI techniques, such as computer vision methods based on deep learning, to assist in detecting EGC 8 . For example, Toshiaki Hirasawa's team 9 developed a CNN diagnostic system that employs the Single Shot MultiBox Detector architecture to process endoscopic images quickly and accurately, achieving an overall sensitivity of 92.2% and a positive predictive value of 30.6%.…”

Section: Introductionmentioning

confidence: 99%

Early gastric cancer detection and lesion segmentation based on deep learning and gastroscopic images

Zhang,

Wang,

Cheng

et al. 2024

Sci Rep

View full text Add to dashboard Cite

Gastric cancer is a highly prevalent disease that poses a serious threat to public health. In clinical practice, gastroscopy is frequently used by medical practitioners to screen for gastric cancer. However, the symptoms of gastric cancer at different stages of advancement vary significantly, particularly in the case of early gastric cancer (EGC). The manifestations of EGC are often indistinct, leading to a detection rate of less than 10%. In recent years, researchers have focused on leveraging deep learning algorithms to assist medical professionals in detecting EGC and thereby improve detection rates. To enhance the ability of deep learning to detect EGC and segment lesions in gastroscopic images, an Improved Mask R-CNN (IMR-CNN) model was proposed. This model incorporates a “Bi-directional feature extraction and fusion module” and a “Purification module for feature channel and space” based on the Mask R-CNN (MR-CNN). Our study includes a dataset of 1120 images of EGC for training and validation of the models. The experimental results indicate that the IMR-CNN model outperforms the original MR-CNN model, with Precision, Recall, Accuracy, Specificity and F1-Score values of 92.9%, 95.3%, 93.9%, 92.5% and 94.1%, respectively. Therefore, our proposed IMR-CNN model has superior detection and lesion segmentation capabilities and can effectively aid doctors in diagnosing EGC from gastroscopic images.

show abstract

Section: Introductionmentioning

confidence: 99%

Early gastric cancer detection and lesion segmentation based on deep learning and gastroscopic images

Zhang,

Wang,

Cheng

et al. 2024

Sci Rep

View full text Add to dashboard Cite

show abstract

“…Gastroscopy uses a thin, flexible tube that is inserted into the stomach, allowing the endoscopist to look directly at the stomach lesions. It is the method of choice for examining gastric lesions and is one of the techniques that must be mastered by gastroenterologists [ 3 ].…”

Section: Introductionmentioning

confidence: 99%

Exploring factors that influence the behavioural intention of medical students to use 3D gastroscopic model to learn how to operate gastroscope using UTAUT Model

Wei

Zhang

et al. 2023

BMC Med Educ

View full text Add to dashboard Cite

Background The application of virtual reality (VR) in gastroscopic operation teaching can be safe and effective, but the advantages can be realized only when students accept and use it. This study aims to identify the factors influencing Chinese clinical medical postgraduates on their intention to use the 3D gastroscopic model constructed based on VR technology using Unified Theory of Acceptance and Use of Technology (UTAUT) model. Students’ demographic factors are also taken into consideration. Methods All methods were carried out in accordance with relevant guidelines. Data were collected from clinical medical postgraduates students in China using stratified sampling. A total of 292 questionnaires including valid responses were used in this study. Data were processed using Amos 24.0 and SPSS 26.0 software and the statistical analysis technique was based on structural equation modeling (SEM). Results The results showed that different from the mediator of home location and year of clinical learning, mediator of gender, university kind and graduate degree did not affect the behavioral intention. In addition, performance expectancy, facilitating condition, and social influence directly and indirectly have effect on behavioral intention. Also, the significance between social influence and performance expectancy, social influence and effort expectancy were verified. Conclusions This study manifested that the proposed framework based on the UTAUT had explanatory power to identify the factors influencing the students’ behavioral intention to use the 3D gastroscopic model constructed based on VR technology. Whereas, an important variable of effort expectancy in the frame of the SEM were not certified, thereby indicating that particular attention should be paid to this variable by universities and teachers before applying 3D gastroscopic model constructed based on VR technology in teaching. Added preparatory work is required such as explaining the basic knowledge of the operating steps of VR model and make students adequately understand its accessibility, which can probably improve the intentions of them to use it. The positive effects of social influence on performance expectancy and effort expectancy we proposed was also verified in this study, which provided a direction for future research.

show abstract

“…The information within the surgical image or video irrespective of the modality used is akin to a digital record of the intervention and it embeds anatomical information, surgical process and event information, data on instrument use and on the interaction between the instruments and the tissue 2 . With the rapid advances seen in artificial intelligence (AI) over the past decade and specifically in computer vision, it is likely that the next generation of interventional capabilities will be built upon AI modules that can extract the information from this rich surgical record and provide computer assisted interventions (CAI) in both perioperative and postoperative settings [3][4][5] (see Fig. 1).…”

Section: A Introductionmentioning

confidence: 99%

Artificial intelligence and automation in endoscopy and surgery

Chadebecq

Lovat

Stoyanov

2022

Nat Rev Gastroenterol Hepatol

View full text Add to dashboard Cite

Modern endoscopic procedures rely on digital technology ranging from high resolution imaging sensors and displays through to electronics connecting configurable illumination and actuation systems for robotic articulation. In addition to enabling more effective diagnostic and therapeutic interventions, the digitization of the procedural toolset also enables video data capture of the internal human anatomy at unprecedented levels. Interventional video data encapsulates functional and structural information about the patients' anatomy, events, activity as well as action log about surgical process. This detailed but difficult to interpret record from endoscopic procedures can potentially be linked to pre-and post-operative records or patient imaging information.Rapid advances in Artificial Intelligence (AI), especially in supervised deep learning, can utilize data from endoscopic procedures to develop systems for assisting procedures leading to computer assisted interventions (CAI). CAI systems can provide a wide range of enhanced capabilities, for example better navigation during procedures, automation of image interpretation or even of robotically assisted tool manipulation. In this review, we summarize the state-of-theart AI for CAI in gastroenterology and surgery.

show abstract

Deep learning for gastroscopic images: computer-aided techniques for clinicians

Cited by 20 publications

References 112 publications

Early gastric cancer detection and lesion segmentation based on deep learning and gastroscopic images

Early gastric cancer detection and lesion segmentation based on deep learning and gastroscopic images

Exploring factors that influence the behavioural intention of medical students to use 3D gastroscopic model to learn how to operate gastroscope using UTAUT Model

Artificial intelligence and automation in endoscopy and surgery

Contact Info

Product

Resources

About