Multi-instance learning for skin biopsy image features recognition

Abstract

In this paper, a multi-instance learning framework is introduced to solve the problem of skin biopsy image features recognition. Previously reported methods for skin surface images were mostly based on color features extraction. They are incapable to be directly applied to skin biopsy image features recognition because biopsy images are often dyed and have obvious inner structures with different textures. Therefore, we regard skin biopsy images as multi-instance samples, whose instances are regions or structures captured by applying Normalized Cut. Texture feature extraction methods are used to express each region as a vectorial expression. Then two multi-instance learning algorithms reported successful in various image retrieval tasks were applied. Nine features were manually selected as target features to evaluate the proposed method on a skin disease diagnosis datasets of 6579 biopsy images from 2010 to 2011. The result showed that the proposed method is effective and medically acceptable.

A sparse Bayesian multi-instance multi-label model for skin biopsy image analysis

“…As shown in our previous work [8], the model that is built upon such methods cannot properly capture the direct medical knowledge and experience for annotating biopsy images. An experienced doctor would annotate an image by directly inspecting some local regions within the image.…”

“…Figure 2 shows the result of Normalized Cut for an skin biopsy image with k = 11. To further express each generated region as a vectorial representation, we apply the method introduced in our previous work [8]. Briefly saying, the feature extraction method performs a waveform transformation for each equal-size block within each region and combine the waveformt transformation coefficiencies to form a 9-ary real vector for each region.…”

“…Recently, Bunte et al [7] proposed a machine learning framework to combine several color space representation methods through a weighting procedure. Zhang et al [8] proposed to convert the skin biopsy image feature recognition problem into a Multi-Instance (MI) learning problem and then solve it by current well-studied MI algorithms, which is the first attempt to tackle the skin biopsy image annotation problem within machine learning framework. In this paper, the authors applied a famous graph cutting algorithm, named Normalized Cut, to generate visual disjoint regions and then apply image feature extraction algorithm for each local region, so as to turn each image into a MI sample.…”

Multi‐instance multi‐label learning for surgical image annotation

“…The images were multilabeled from a set of 15 annotation terms, and the instances were represented by feature vectors extracted from local regions. In a similar work, a multi‐instance (MI) learner based on the citation‐kNN algorithm was trained for each annotation term . However, the learner discarded the correlation between the target labels.…”

“…In a similar work, a multi-instance (MI) learner based on the citation-kNN algorithm was trained for each annotation term. 21 However, the learner discarded the correlation between the target labels.…”