Computationally Derived Image Signature of Stromal Morphology Is Prognostic of Prostate Cancer Recurrence Following Prostatectomy in African American Patients

Bhargava, Hersh K.; Leo, Patrick; Elliott, Robin; Janowczyk, Andrew; Whitney, Jon; Gupta, Sanjay; Fu, Pingfu; Yamoah, Kosj; Khani, Francesca; Robinson, Brian D.; Rebbeck, Timothy R.; Feldman, Michael D.; Lal, Priti; Madabhushi, Anant

doi:10.1158/1078-0432.ccr-19-2659

Cited by 44 publications

(40 citation statements)

References 27 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Multi-stage feature-based classification approaches rely on the accurate segmentation of morphological structures such as nuclei (30)(31)(32), collagen fibers (33,34), vessels (14,35), or in our case, prostate glands (21,36). This is typically achieved in one of two ways: (1) Direct DLbased segmentation methods (37)(38)(39)(40) that require manually annotated training datasets, which are especially tedious and difficult to obtain in 3D (Fig.…”

Section: Introductionmentioning

confidence: 99%

Prostate Cancer Risk Stratification via Nondestructive 3D Pathology with Deep Learning–Assisted Gland Analysis

et al. 2021

Self Cite

View full text Add to dashboard Cite

are co-founders and shareholders of Lightspeed Microscopy Inc. A. Madabhushi is an equity holder in Elucid Bioimaging and in Inspirata Inc. In addition, he has served as a scientific advisory board member for Inspirata Inc, Astrazeneca, Bristol Meyers-Squibb and Merck. Currently, A. Madabhushi serves on the advisory board of Aiforia Inc. He also has sponsored research agreements with Philips, AstraZeneca, Boehringer-Ingelheim and Bristol Meyers-Squibb. A. Madabhushi's technology has been licensed to Elucid Bioimaging. He is also involved in a NIH U24 grant with PathCore Inc, and 3 different R01 grants with Inspirata Inc.Research.

show abstract

Section: Introductionmentioning

confidence: 99%

Prostate Cancer Risk Stratification via Nondestructive 3D Pathology with Deep Learning–Assisted Gland Analysis

et al. 2021

Self Cite

View full text Add to dashboard Cite

show abstract

“…Zhao et al 39 reported that the PD-L2 is a prognostic biomarker for PCa based on patients, and they also reported that the in ltration of T cells and Macrophages are increased in the poor outcome group, which is also consistent with our work that M2 macrophages link with the unfavorable prognosis, while the immunocytes and clinical feature-based Combined could distinguish the different ending of recurrence (AUC = 0.91) 40 . Bhargava et al 41 illustrated an African-American speci cally automated stromal classi er, which has the potential to substantially improve the accuracy of prognosis and risk strati cation. Yang et al 42 established a 28hypoxia-related-gene prognostic classi er for localized PCa, which could predict the biochemical recurrence and metastasis events.…”

Section: Discussionmentioning

confidence: 99%

Molecular subtyping and prognostic assessment of prostate cancer based on consensus genes

Meng

Guo

Asan

et al. 2021

Preprint

View full text Add to dashboard Cite

Prostate cancer (PCa) is one of the most common malignancies in men all over the world. We performed molecular subtyping and prognostic assessment based on consensus genes in patients with PCa. Five cohorts containing 1,046 PCa patients with RNA expression profiles and recorded clinical follow-up information were included. Univariate, multivariate Cox regression analysis and the Least Absolute Shrinkage and Selection Operator (LASSO) Cox regression was used to select prognostic genes and establish the signature. Immunohistochemistry staining, cell proliferation, migration and invasion assay were used to assess the biological function of key genes. 39 intersecting consensus prognostic genes from five independent cohorts were identified (P < 0.05). Subsequently, an eleven-consensus-gene-based classifier was established. Besides, multivariate Cox regression analyses proved that the classifier served as an independent indicator of recurrence-free survival in three of five cohorts. Combined receiver operating character (ROC) achieved synthesized effects by combining the classifier with clinicopathological features in four of five cohorts. SRD5A2 inhibits the cell proliferation, while ITGA11 promote cell migration and invasion, of which might through PI3K/AKT signaling pathway. To conclude, we establish and validate an eleven-consensus-gene-based classifier, which adds prognostic value to the currently available staging system.

show abstract

“…In cancer, race-specific variations in occurrence and frequency of genomic aberrations have been reported (122). Work by Bhargava and colleagues has in fact shown that race-specific differences exist even at the level of tissue morphology-and so do differences in disease aggressiveness-between Caucasian and African American men with prostate cancer (123). But existing datasets that are commonly used to train and test AI models in cancer are still inherently biased toward certain racial and ethnic groups.…”

Section: Current Challenges and Future Perspectivesmentioning

confidence: 99%

Artificial Intelligence in Cancer Research and Precision Medicine

Bhinder

Gilvary²,

Madhukar³

et al. 2021

Cancer Discovery

251

121

View full text Add to dashboard Cite

Convolutional neural networks (CNN) have been the most popular deep learning architectures used for image classification in cancer ( Fig. 1 ). CNNs apply a series of nonlinear transformations to structured data (such as raw pixels of an image) to learn relevant features automatically, unlike conventional machine learning models that frequently require manual feature curation. On the fl ip side, it is diffi cult to tell what features are learned by the CNNs, making them what many have referred to as a "black box." One consequence is that images used for CNNs should be carefully prepro cessed to reduce the risk that the model learns from image artifacts. There are two major approaches for CNN models; one is transfer learning that uses images from a large collection of natural objects (such as in ImageNet) to train the initial layers of a model (where the model learns to identify general features such as shapes and edges) and then uses the diseasespecifi c data to fi ne-tune the training parameters in the last layers. The second variation of CNN is based on an autoencoder where the model learns background features from a subset of representative images and encodes a compressed representation of the basic features later used to initialize the CNN. In the CAMELYON16 Challenge-a crowdsourced competition to identify and classify lymph node metastasis in patients with breast cancer from whole slide images (WSI) of hematoxylin and eosin (H&E)-stained tumors-25 of the 32 submitted algorithms were CNNs, and the top 5 classifi cation models were exclusively based on transfer learning, which were GoogLeNet, ResNet, VGG-16 ( 2 ). Khosravi and colleagues trained and tested several state-of-the-art deep learning models to classify WSI from H&E-stained tumor tissues of The Cancer Genome Atlas (TCGA) cohort and reported on the

show abstract

Computationally Derived Image Signature of Stromal Morphology Is Prognostic of Prostate Cancer Recurrence Following Prostatectomy in African American Patients

Cited by 44 publications

References 27 publications

Prostate Cancer Risk Stratification via Nondestructive 3D Pathology with Deep Learning–Assisted Gland Analysis

Prostate Cancer Risk Stratification via Nondestructive 3D Pathology with Deep Learning–Assisted Gland Analysis

Molecular subtyping and prognostic assessment of prostate cancer based on consensus genes

Artificial Intelligence in Cancer Research and Precision Medicine

Contact Info

Product

Resources

About