Vision Transformers for Small Histological Datasets Learned Through Knowledge Distillation

Kanwal, Neel; Eftestøl, Trygve; Khoraminia, Farbod; Zuiverloon, Tahlita C.M.; Engan, Kjersti

doi:10.1007/978-3-031-33380-4_13

Cited by 10 publications

(10 citation statements)

References 22 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…To enhance the generalizability of DL models, incremental learning may be used to tune models on other public datasets, such as PICTURE and PROSTA-TEx [28]. To increase the precision of PCa grading, future studies can concentrate on integrating vision transformers (ViTs) [12]. Perhaps in conjunction with data from other modalities, mpMRI can potentially improve PCa diagnosis substantially using DL models.…”

Section: Discussionmentioning

confidence: 99%

“…The future of automated PCa grading DL models will benefit from vision transformers (ViTs) [4] and their multi-attention [36] to the spatial correlation of mpMRI images. This work can be further extended by combining ViTs and CNNs for PCa grading tasks, possibly in combination with data from different modalities.…”

Section: Limitations and Future Workmentioning

confidence: 99%

“…DL algorithms can accurately detect cancer by identifying specific features in mpMRI images associated with PCa, such as tumor size and location. Computer-aided (CAD) tools using DL methods can provide decision support to clinicians, helping them focus on regions of interest (ROIs) and optimizing time efficiency in the diagnostic procedure [9], [12]. CAD tools also provide personalized treatment strategies based on the grade of cancer.…”

Section: Introductionmentioning

confidence: 99%

“…Deep learning (DL) approaches are becoming popular in biomedical image analysis [3,4]. DL approaches are promising due to their potential to identify complex patterns and hidden representations in the data.…”

Section: Introductionmentioning

confidence: 99%

“…DL approaches are promising due to their potential to identify complex patterns and hidden representations in the data. DL models, such as deep neural networks (DNNs) have shown great potential in accurately detecting and classifying medical images, which can help improve the accuracy of diagnosis and treatment planning [4,5]. Computeraided (CAD) systems using DL approaches can provide decision support to doctors by localizing regions of interest and saving their time in collecting second opinions [6,7].…”

Section: Introductionmentioning

confidence: 99%

See 4 more Smart Citations

Automated Diagnosis of Prostate Cancer Using mpMRI Images: A Deep Learning Approach for Clinical Decision Support

Gavade¹,

Nerli²,

Kanwal

et al. 2023

Computers

Self Cite

View full text Add to dashboard Cite

Prostate cancer (PCa) is a significant health concern for men worldwide, where early detection and effective diagnosis can be crucial for successful treatment. Multiparametric magnetic resonance imaging (mpMRI) has evolved into a significant imaging modality in this regard, which provides detailed images of the anatomy and tissue characteristics of the prostate gland. However, interpreting mpMRI images can be challenging for humans due to the wide range of appearances and features of PCa, which can be subtle and difficult to distinguish from normal prostate tissue. Deep learning (DL) approaches can be beneficial in this regard by automatically differentiating relevant features and providing an automated diagnosis of PCa. DL models can assist the existing clinical decision support system by saving a physician’s time in localizing regions of interest (ROIs) and help in providing better patient care. In this paper, contemporary DL models are used to create a pipeline for the segmentation and classification of mpMRI images. Our DL approach follows two steps: a U-Net architecture for segmenting ROI in the first stage and a long short-term memory (LSTM) network for classifying the ROI as either cancerous or non-cancerous. We trained our DL models on the I2CVB (Initiative for Collaborative Computer Vision Benchmarking) dataset and conducted a thorough comparison with our experimental setup. Our proposed DL approach, with simpler architectures and training strategy using a single dataset, outperforms existing techniques in the literature. Results demonstrate that the proposed approach can detect PCa disease with high precision and also has a high potential to improve clinical assessment.

show abstract

Section: Discussionmentioning

confidence: 99%

Section: Limitations and Future Workmentioning

confidence: 99%