Evolutionary deep feature selection for compact representation of gigapixel images in digital pathology

Bidgoli, Azam Asilian; Rahnamayan, Shahryar; Dehkharghanian, Taher; Riasatian, Abtin; Kalra, Shivam; Zaveri, Manit; Campbell, Clinton J. V.; Parwani, Anil V.; Pantanowitz, Liron; Tizhoosh, Hamid R.

doi:10.1016/j.artmed.2022.102368

Cited by 9 publications

(8 citation statements)

References 39 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“… 7 Combined with the high accuracy of SAMPLER classifiers, the biological meaningfulness of these attention maps show that SAMPLER representations successfully encode the distinct morphologies within WSIs. An advantage of SAMPLER over previous statistical models that encode each WSI using mean or median values of deep learning features 7 , 52 is that SAMPLER yields feature-decomposable attention maps. Mean/median approaches have achieved high accuracies in certain tasks, 7 , 52 but they do not output attention maps, which limits their interpretability.…”

Section: Discussionmentioning

confidence: 99%

SAMPLER: unsupervised representations for rapid analysis of whole slide tissue images

Mukashyaka,

Sheridan,

Foroughi pour

et al. 2024

eBioMedicine

View full text Add to dashboard Cite

Section: Discussionmentioning

confidence: 99%

SAMPLER: unsupervised representations for rapid analysis of whole slide tissue images

Mukashyaka,

Sheridan,

Foroughi pour

et al. 2024

eBioMedicine

View full text Add to dashboard Cite

“…Hence, averaging feature vector gives a soft vector. This approach has been successfully used for compact WSI representation [18].…”

Section: Methodsmentioning

confidence: 99%

Deep modality association learning using histopathology images and immune cell sequencing data

Alsaafin

Babaie

Tizhoosh

2023

Medical Imaging 2023: Digital and Computational Pathology

View full text Add to dashboard Cite

Nowadays, the availability of different types of biomedical digital data offers many opportunities to investigate the relationships between the different modalities and thus develop a more comprehensive understanding of complex diseases such as cancer. In this paper, we propose a multi-modal model, called deep modality association learning (DMAL), that maps immune cell sequencing patterns to morphological tissue features of whole slide imageds (WSIs) in an embedding space. Useful information is extracted from T-cell receptor (TCR) sequences to guide the training process. DMAL maps the TCR features to the morphology features in histopathology images, which in turn enables the model to learn the association features between the two modalities. The discrimination power of the WSI-TCR association features has been assessed by classifying samples with different cancer subtypes. The conducted experiments have shown that DMAL generates more discriminative features compared to features obtained from single-modal data. In addition, DMAL has been utilized to predict TCR information from histopathology image representations without the need to have the actual TCR sequencing data.

show abstract

“…In [29], the proposed approach involves the use of a multi‐objective evolutionary algorithm for feature selection in digital pathology. This process reduces the feature vector to 11,000 times smaller than the initial feature vector.…”

Section: Relevant Studiesmentioning

confidence: 99%

“…This storage requirement varies based on the number of neurons in the fully connected layers. Many studies have demonstrated that the removal of irrelevant features complements the improvement of deep learning model performance, especially for high-dimensional data [29][30][31]. Therefore, in this case, feature selection can be considered as an optimization problem with two conflicting objectives: maximizing classification accuracy and minimizing the number of features simultaneously.…”

Section: Introductionmentioning

confidence: 99%

“…Despite the numerous advantages of deep learning, including remarkable prediction accuracy, their ability to model complex processes, generate structured high‐level feature representations, and their resistance to temporal or spatial variations [28], it is important to note that the effectiveness of these models is highly dependent on network design and proper training. Non‐optimal choices, such as increasing network depth or using improper values for network hyperparameters, can lead to the production of irrelevant and redundant features [29]. Therefore, when extracting features from these networks, the obtained feature vector may contain irrelevant features.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

A comparative study on deep feature selection methods for skin lesion classification

Golnoori,

Zamani Boroujeni,

Monadjemi

2023

IET Image Processing

View full text Add to dashboard Cite

Melanoma, a widespread and hazardous form of cancer, has prompted researchers to prioritize dermoscopic image‐based algorithms for classifying skin lesions. Recently, there has been a growing trend in using pre‐trained convolutional neural networks for detecting skin lesions. However, the features extracted from these classifiers may include irrelevant elements, emphasizing the importance of implementing effective feature selection methods. Nevertheless, there has not been a comprehensive study on feature selection methods to enhance the performance of skin lesion detection to date. To identify the most efficient methods, a diverse set of feature selection techniques, including filter, wrapper, embedded, and dimensionality reduction, were applied to images from two well‐known datasets, namely ISIC 2017 and ISIC 2018. According to the results, models trained with features chosen by wrapper techniques outperformed those trained with features chosen by filter and embedded methods. Achieving an accuracy of 0.8333 and an F1‐Score of 0.8291 for the ISIC 2017 dataset, and an F1‐Score of 0.9324 and an accuracy of 0.9350 for the ISIC 2018 dataset, the classification using features obtained via the GWO feature selection technique performed the best.

show abstract

Evolutionary deep feature selection for compact representation of gigapixel images in digital pathology

Cited by 9 publications

References 39 publications

SAMPLER: unsupervised representations for rapid analysis of whole slide tissue images

SAMPLER: unsupervised representations for rapid analysis of whole slide tissue images

Deep modality association learning using histopathology images and immune cell sequencing data

A comparative study on deep feature selection methods for skin lesion classification

Contact Info

Product

Resources

About