Feature fusion of Raman chemical imaging and digital histopathology using machine learning for prostate cancer detection

Doherty, Trevor; McKeever, Susan; Al-Attar, Nebras; Murphy, Tiarnan; Aura, Claudia; Rahman, Arman; O’Neill, Amanda; Finn, Stephen; Kay, Elaine W.; Watson, R. William G.; Gowen, Aoife; Jackman, Patrick

doi:10.1039/d1an00075f

Cited by 13 publications

(8 citation statements)

References 72 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…110 Using a binary classification model, a sensitivity of 73.8% and a specificity of 88.1% were predicted for the G3/G4 classification, while a sensitivity of 54.1% and specificity of 84.7% were obtained using only digital histopathology. 110 Head and Neck Cancer. Zhang et al distinguished between normal (n = 34) and laryngeal squamous cell carcinoma (n = 44) using SRH imaging at only two wavenumbers (2845 and 2930 cm −1 ) in addition to SHG imaging.…”

Section: ■ Raman Spectroscopy For Diagnosticsmentioning

confidence: 95%

“…Doherty et al reported that a combination of digital histopathology and spontaneous Raman imaging may advance the diagnosis of prostate cancer because of incorporating morphological and biochemical information . Using a binary classification model, a sensitivity of 73.8% and a specificity of 88.1% were predicted for the G3/G4 classification, while a sensitivity of 54.1% and specificity of 84.7% were obtained using only digital histopathology …”

Section: Raman Spectroscopy For Diagnosticsmentioning

confidence: 99%

See 1 more Smart Citation

Diagnostics and Therapy Assessment Using Label-Free Raman Imaging

El-Mashtoly

Gerwert

2021

Anal. Chem.

View full text Add to dashboard Cite

Section: ■ Raman Spectroscopy For Diagnosticsmentioning

confidence: 95%

Section: Raman Spectroscopy For Diagnosticsmentioning

confidence: 99%

Diagnostics and Therapy Assessment Using Label-Free Raman Imaging

El-Mashtoly

Gerwert

2021

Anal. Chem.

View full text Add to dashboard Cite

“…Therefore, the instruments need to perform quick data assessments and analysis. Here, first approaches report on the implementation of automated and machine-learning based classification and readout options [ 273 , 274 , 275 ], which will be an inevitable task in the future, especially when multimodal imaging approaches and conventional imaging tools will be combined in the surgery room.…”

Section: Challenges and Limitationsmentioning

confidence: 99%

Raman Imaging and Fluorescence Lifetime Imaging Microscopy for Diagnosis of Cancer State and Metabolic Monitoring

Becker

Janssen

Layland

et al. 2021

Cancers

View full text Add to dashboard Cite

Hurdles for effective tumor therapy are delayed detection and limited effectiveness of systemic drug therapies by patient-specific multidrug resistance. Non-invasive bioimaging tools such as fluorescence lifetime imaging microscopy (FLIM) and Raman-microspectroscopy have evolved over the last decade, providing the potential to be translated into clinics for early-stage disease detection, in vitro drug screening, and drug efficacy studies in personalized medicine. Accessing tissue- and cell-specific spectral signatures, Raman microspectroscopy has emerged as a diagnostic tool to identify precancerous lesions, cancer stages, or cell malignancy. In vivo Raman measurements have been enabled by recent technological advances in Raman endoscopy and signal-enhancing setups such as coherent anti-stokes Raman spectroscopy or surface-enhanced Raman spectroscopy. FLIM enables in situ investigations of metabolic processes such as glycolysis, oxidative stress, or mitochondrial activity by using the autofluorescence of co-enzymes NADH and FAD, which are associated with intrinsic proteins as a direct measure of tumor metabolism, cell death stages and drug efficacy. The combination of non-invasive and molecular-sensitive in situ techniques and advanced 3D tumor models such as patient-derived organoids or microtumors allows the recapitulation of tumor physiology and metabolism in vitro and facilitates the screening for patient-individualized drug treatment options.

show abstract

“…The core of this algorithm is a feature fusion algorithm that can automatically adjust the proportion of high-level DL features and traditional features ( 20 ). Similarly, the fusion of digital histopathology and Raman chemical imaging modalities has the potential to improve the binary classification of prostate cancer pathology images by integrating both morphological and biochemical information across data sources ( 21 ). Most recently, to mine more information from different radiomics data in multicenter studies and identify personalized predictive and/or prognostic models to improve the reproducibility, an image biomarker standardization initiative (IBSI) was introduced for the standardization of radiomic features ( 22 ), and some radiomics computational frameworks were developed to comply IBSI and allow users to complete the whole radiomics workflow within the same software, simplifying the radiomics process, such as the matRadiomics software ( 23 ).…”

Section: Introductionmentioning

confidence: 99%

Fusing hand-crafted and deep-learning features in a convolutional neural network model to identify prostate cancer in pathology images

Huang

Zhang

et al. 2022

Front. Oncol.

View full text Add to dashboard Cite

Prostate cancer can be diagnosed by prostate biopsy using transectal ultrasound guidance. The high number of pathology images from biopsy tissues is a burden on pathologists, and analysis is subjective and susceptible to inter-rater variability. The use of machine learning techniques could make prostate histopathology diagnostics more precise, consistent, and efficient overall. This paper presents a new classification fusion network model that was created by fusing eight advanced image features: seven hand-crafted features and one deep-learning feature. These features are the scale-invariant feature transform (SIFT), speeded up robust feature (SURF), oriented features from accelerated segment test (FAST) and rotated binary robust independent elementary features (BRIEF) (ORB) of local features, shape and texture features of the cell nuclei, the histogram of oriented gradients (HOG) feature of the cavities, a color feature, and a convolution deep-learning feature. Matching, integrated, and fusion networks are the three essential components of the proposed deep-learning network. The integrated network consists of both a backbone and an additional network. When classifying 1100 prostate pathology images using this fusion network with different backbones (ResNet-18/50, VGG-11/16, and DenseNet-121/201), we discovered that the proposed model with the ResNet-18 backbone achieved the best performance in terms of the accuracy (95.54%), specificity (93.64%), and sensitivity (97.27%) as well as the area under the receiver operating characteristic curve (98.34%). However, each of the assessment criteria for these separate features had a value lower than 90%, which demonstrates that the suggested model combines differently derived characteristics in an effective manner. Moreover, a Grad-CAM++ heatmap was used to observe the differences between the proposed model and ResNet-18 in terms of the regions of interest. This map showed that the proposed model was better at focusing on cancerous cells than ResNet-18. Hence, the proposed classification fusion network, which combines hand-crafted features and a deep-learning feature, is useful for computer-aided diagnoses based on pathology images of prostate cancer. Because of the similarities in the feature engineering and deep learning for different types of pathology images, the proposed method could be used for other pathology images, such as those of breast, thyroid cancer.

show abstract

Feature fusion of Raman chemical imaging and digital histopathology using machine learning for prostate cancer detection

Abstract: The diagnosis of prostate cancer is challenging due to the heterogeneity of its presentations, leading to the over diagnosis and treatment of non-clinically important disease. Accurate diagnosis can directly benefit...

Cited by 13 publications

References 72 publications

Diagnostics and Therapy Assessment Using Label-Free Raman Imaging

Diagnostics and Therapy Assessment Using Label-Free Raman Imaging

Raman Imaging and Fluorescence Lifetime Imaging Microscopy for Diagnosis of Cancer State and Metabolic Monitoring

Fusing hand-crafted and deep-learning features in a convolutional neural network model to identify prostate cancer in pathology images

Contact Info

Product

Resources

About