Kvasir-Capsule, a video capsule endoscopy dataset

Abstract

Artificial intelligence (AI) is predicted to have profound effects on the future of video capsule endoscopy (VCE) technology. The potential lies in improving anomaly detection while reducing manual labour. Existing work demonstrates the promising benefits of AI-based computer-assisted diagnosis systems for VCE. They also show great potential for improvements to achieve even better results. Also, medical data is often sparse and unavailable to the research community, and qualified medical personnel rarely have time for the tedious labelling work. We present Kvasir-Capsule, a large VCE dataset collected from examinations at a Norwegian Hospital. Kvasir-Capsule consists of 117 videos which can be used to extract a total of 4,741,504 image frames. We have labelled and medically verified 47,238 frames with a bounding box around findings from 14 different classes. In addition to these labelled images, there are 4,694,266 unlabelled frames included in the dataset. The Kvasir-Capsule dataset can play a valuable role in developing better algorithms in order to reach true potential of VCE technology.

DeepSynthBody: the beginning of the end for data deficiency in medicine

“…However, we soon realized a huge lack of medical data to develop good ML models in the domain for various reasons, increasing the importance of the first two steps in Figure 1 Therefore, we have studied how datasets should be collected, composed, and published as open datasets. Within the three years of Ph.D. time, a total of seven datasets [23,24,25,26,27,28,29] were successfully collected and published. In these datasets, medical experts labeled or annotated data (Step II), but not all the datasets because the annotation process is costly and time-consuming.…”

“…For the cardiology branch, an ML-based ECG analysis system [41] was researched and implemented. Moreover, all the dataset papers [23,24,25,26,27,28,29] introduced ML models as baseline experiments which can be considered initial models for developing CAD systems.…”

“…Moreover, these datasets are the main assets for initiating the novel concept and the corresponding framework, DeepSynthBody, introduced in this thesis. Different types of real medical datasets [23,24,25,26,27,28,29] were collected and published to the research community with the baseline experiments under this thesis to accomplish the sub-objective I,. All the datasets contribute to designing ML models for CAD systems (sub-objective I) because of the baseline experiments introduced in every dataset paper.…”

“…Therefore, collecting medical data is identified as a key step for the thesis. As a result, several datasets [23,24,25,26,27,28,29] were published. More details about these datasets are discussed in Section 3.1.1.…”

Artificial intelligence, capsule endoscopy, databases, and the Sword of Damocles

“…Especially in the field of capsule endoscopy (CE), where imaging data are readily available, it remains to be determined who will plough through the images, delineate/annotate and comment on regions of interest, and make sure that DL training is performed with high-quality material. Considering this, putting in a substantial amount of human effort (including personal) 1 , we set off to create a series of respective CE databases, i. e., KID, CAD-CAP, and Kvasir Capsule 4 5 6 for the benefit of computer scientists, at the expense of effort by ourselves and colleagues. Although they are enriched by and enlarged with CE images from different manufacturers, the diverse databases contain numerous classes of gastrointestinal normal and abnormal findings that have been prepared in various way.…”