Addressing the Inevitable Imprecision: Multiple Instance Learning for Hyperspectral Image Analysis

Jiao, Changzhe; Du, Xiaomeng; Zare, Alina

doi:10.1007/978-3-030-38617-7_6

Cited by 5 publications

(2 citation statements)

References 56 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The term MIL was coined in the 1990′s [14] and the associated methods have been extensively used in diverse fields such as computational pathology, [12][13][14][15][16][17] natural language processing, [18,19] and hyperspectral remote sensing. [20][21][22][23] The MIL problem formulation considers the data set as bags (in our case, the entire hyperspectral MSI images) composed of instances. In this work, we consider individual MSI pixels as the instances; however, it would be equally feasible to consider a patch of pixels as an instance.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Markedly Enhanced Analysis of Mass Spectrometry Images Using Weakly Supervised Machine Learning

Gardner,

Winkler,

Bamford

et al. 2024

Small Methods

View full text Add to dashboard Cite

Supervised and unsupervised machine learning algorithms are routinely applied to time‐of‐flight secondary ion mass spectrometry (ToF‐SIMS) imaging data and, more broadly, to mass spectrometry imaging (MSI). These algorithms have accelerated large‐scale, single‐pixel analysis, classification, and regression. However, there is relatively little research on methods suited for so‐called weakly supervised problems, where ground‐truth class labels exist at the image level, but not at the individual pixel level. Unsupervised learning methods are usually applied to these problems. However, these methods cannot make use of available labels. Here a novel method specifically designed for weakly supervised MSI data is presented. A dual‐stream multiple instance learning (MIL) approach is adapted from computational pathology that reveals the spatial‐spectral characteristics distinguishing different classes of MSI images. The method uses an information entropy‐regularized attention mechanism to identify characteristic class pixels that are then used to extract characteristic mass spectra. This work provides a proof‐of‐concept exemplification using printed ink samples imaged by ToF‐SIMS. A second application‐oriented study is also presented, focusing on the analysis of a mixed powder sample type. Results demonstrate the potential of the MIL method for broader application in MSI, with implications for understanding subtle spatial‐spectral characteristics in various applications and contexts.

show abstract

Section: Introductionmentioning

confidence: 99%

“…Such data are referred to as weakly labeled. [ 12–23 ] As this is a common scenario, ML methods specifically designed for weakly labeled MSI data are necessary, especially given the importance of MSI for these applications. This is the focus of the current work.…”

Section: Introductionmentioning

confidence: 99%