Dimensionality reduction of mass spectrometry imaging data using autoencoders

Thomas, Spencer A.; Race, Alan; Steven, Rory T.; Gilmore, Ian S.; Bunch, Josephine

doi:10.1109/ssci.2016.7849863

Cited by 47 publications

(35 citation statements)

References 42 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Last, we discuss ways to interpret the learned model and analyze if biologically plausible effects are visible, a crucial step for applying it to tumor diagnostic. Deep learning has been introduced to IMS data prior to this work, but with a focus on unsupervised dimension reduction methods, see (Thomas et al, 2016) where autoencoders were used to reduce rat brain IMS data. Moreover, (Inglese et al, 2017) introduced a neural network based dimension reduction to find metabolic regions within tumors.…”

Section: Introductionmentioning

confidence: 99%

Deep learning for tumor classification in imaging mass spectrometry

et al. 2017

View full text Add to dashboard Cite

Motivation: Tumor classification using Imaging Mass Spectrometry (IMS) data has a high potential for future applications in pathology. Due to the complexity and size of the data, automated feature extraction and classification steps are required to fully process the data. Deep learning offers an approach to learn feature extraction and classification combined in a single model. Commonly these steps are handled separately in IMS data analysis, hence deep learning offers an alternative strategy worthwhile to explore. Results: Methodologically, we propose an adapted architecture based on deep convolutional networks to handle the characteristics of mass spectrometry data, as well as a strategy to interpret the learned model in the spectral domain based on a sensitivity analysis. The proposed methods are evaluated on two challenging tumor classification tasks and compared to a baseline approach. Competitiveness of the proposed methods are shown on both tasks by studying the performance via cross-validation. Moreover, the learned models are analyzed by the proposed sensitivity analysis revealing biologically plausible effects as well as confounding factors of the considered task. Thus, this study may serve as a starting point for further development of deep learning approaches in IMS classification tasks. Source Code: https://gitlab.informatik.uni-bremen.de/digipath/Deep Learning for Tumor Classification in IMS Data: https://seafile.zfn.uni-bremen.de/d/85c915784e/

show abstract

Section: Introductionmentioning

confidence: 99%

Deep learning for tumor classification in imaging mass spectrometry

et al. 2017

View full text Add to dashboard Cite

show abstract

“…119 Autoencoders have also been useful for unsupervised nonlinear dimensionality reduction of imaging data by reducing each pixel one at a time to its core features. 120 Once the size of data has been reduced, it can be more easily processed in subsequent steps of the processing pipeline.…”

Section: Resultsmentioning

confidence: 99%

Mass Spectrometry Imaging: A Review of Emerging Advancements and Future Insights

et al. 2017

View full text Add to dashboard Cite

“…Activation functions such as ReLU, σ (z) = max(0, z), are popular in convolutional neural networks due to their success in image analysis problems [16]. Sigmoid activation functions, σ (z) = 1 + e −z −1 , have also been shown to be useful in reduction and visualization of complex and high dimensional data [22], and identification of low dimensional patterns in the data for segmentation and classification tasks [16,18]. Here we compare both ReLU and sigmoid activation functions for reduction of the CMR data.…”

Section: Deep Autoencodermentioning

confidence: 99%