Abstract
This paper introduces a supervised learning method for finding diagnostic regions of interest in histopathological images. The method is based on the cognitive process of visual selection of relevant regions that arises during a pathologist's image examination. The proposed strategy emulates the interaction of the visual cortex areas V1, V2 and V4, being the V1 cortex responsible for assigning local levels of relevance to visual inputs while the V2 cortex gathers together these small regions according to some weights modulated by the V4 cortex, which stores some learned rules. This novel strategy can be considered as a complex mix of "bottom-up" and "top-down" mechanisms, integrated by calculating a unique index inside each region. The method was evaluated on a set of 338 images in which an expert pathologist had drawn the Regions of Interest. The proposed method outperforms two state-of-the-art methods devised to determine Regions of Interest (RoIs) in natural images. The quality gain with respect to an adaptated Itti's model which found RoIs was 3.6 dB in average, while with respect to the Achanta's proposal was 4.9 dB.
Highlights
A typical pathology laboratory examines more than 100 microscopical slides per day [1], a scenario in which its workflow is based on the interaction of the pathologists with a conventional microscope
This standard should deal with three main questions: 1) What quality level meets the minimal diagnosis conditions, avoiding wrong diagnosis decisions?, 2) What quality level is needed for accurate diagnoses?
We used them as the base line because they emulate the visual system, even though they are not devised to detect relevancy in medical images, these two models were: 1) the Itti’s model [18,29], one of the most popular algorithms to find Regions of Interest (RoIs) in an image
Summary
A typical pathology laboratory examines more than 100 microscopical slides per day [1], a scenario in which its workflow is based on the interaction of the pathologists with a conventional microscope. The lack of standardized criteria to preserve data reliability -from the early capturing process to the final interpretation-, limits the routine used of virtual microscopy techniques, in despite of its obvious technical advantages, namely, second opinions, team work, image annotation, deterioration-free digital storing. Such a standard should provide a robust frame, allowing the pathologists to achieve proper diagnoses, since it should garantee that the image data will be free of any artifact introduced during the slide preparation, digitization, transmission or visualization. Image regions, considered as relevant, require in general higher quality levels
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
