scholarly journals Using Histopathology Images to Predict Chromosomal Instability in Breast Cancer: A Deep Learning Approach

2020 ◽  
Author(s):  
Zhuoran Xu ◽  
Akanksha Verma ◽  
Uska Naveed ◽  
Samuel Bakhoum ◽  
Pegah Khosravi ◽  
...  
Keyword(s):  
Breast Cancer ◽  
Deep Learning ◽  
Spatial Heterogeneity ◽  
Cancer Patients ◽  
Gold Standard ◽  
Chromosomal Instability ◽  
Human Cancer ◽  
Breast Cancer Subtype ◽  
The Cancer Genome Atlas ◽  
Test Set

Chromosomal instability (CIN) is a hallmark of human cancer that involves mis-segregation of chromosomes during mitosis, leading to aneuploidy and genomic copy number heterogeneity. CIN is a prognostic marker in a variety of cancers, yet, gold-standard experimental assessment of chromosome mis-segregation is difficult in the routine clinical setting. As a result, CIN status is not readily testable for cancer patients in such setting. On the other hand, the gold-standard for cancer diagnosis and grading, histopathological examinations, are ubiquitously available. In this study, we sought to explore whether CIN status can be predicted using hematoxylin and eosin (H&E) histology in breast cancer patients. Specifically, we examined whether CIN, defined using a genomic aneuploidy burden approach, can be predicted using a deep learning-based model. We applied transfer learning on convolutional neural network (CNN) models to extract histological features and trained a multilayer perceptron (MLP) after aggregating patch features obtained from whole slide images. When applied to a breast cancer cohort of 1,010 patients (Training set: n=858 patients, Test set: n=152 patients) from The Cancer Genome Atlas (TCGA) where 485 patients have high CIN status, our model accurately classified CIN status, achieving an area under the curve (AUC) of 0.822 with 81.2% sensitivity and 68.7% specificity in the test set. Patch-level predictions of CIN status suggested intra-tumor spatial heterogeneity within slides. Moreover, presence of patches with high predicted CIN score within an entire slide was more predictive of clinical outcome than the average CIN score of the slide, thus underscoring the clinical importance of spatial heterogeneity. Overall, we demonstrated the ability of deep learning methods to predict CIN status based on histopathology slide images. Our model is not breast cancer subtype specific and the method can be potentially extended to other cancer types.

scholarly journals Hybrid Approach of Relation Network and Localized Graph Convolutional Filtering for Breast Cancer Subtype Classification

2018 ◽  
Author(s):  
Sungmin Rhee ◽  
Seokjun Seo ◽  
Sun Kim
Keyword(s):  
Breast Cancer ◽  
Deep Learning ◽  
Cancer Biology ◽  
Current Knowledge ◽  
Human Cancer ◽  
Breast Cancer Subtype ◽  
Hybrid Approach ◽  
Network Biology ◽  
Learning Technologies ◽  
Cancer Data

Network biology has been successfully used to help reveal complex mechanisms of disease, especially cancer. On the other hand, network biology requires in-depth knowledge to construct disease-specific networks, but our current knowledge is very limited even with the recent advances in human cancer biology. Deep learning has shown an ability to address the problem like this. However, it conventionally used grid-like structured data, thus application of deep learning technologies to the human disease subtypes is yet to be explored. To overcome the issue, we propose a hybrid model, which integrates two key components 1) graph convolution neural network (graph CNN) and 2) relation network (RN). Experimental results on synthetic data and breast cancer data demonstrate that our proposed method shows better performances than existing methods.

scholarly journals Human Plasma Metabolomics for Biomarker Discovery: Targeting the Molecular Subtypes in Breast Cancer

Cancers ◽  
2021 ◽  
Vol 13 (1) ◽  
pp. 147
Author(s):  
Leticia Díaz-Beltrán ◽  
Carmen González-Olmedo ◽  
Natalia Luque-Caro ◽  
Caridad Díaz ◽  
Ariadna Martín-Blázquez ◽  
...  
Keyword(s):  
Breast Cancer ◽  
Cancer Patients ◽  
Biomarker Discovery ◽  
Molecular Subtypes ◽  
Breast Cancer Subtype ◽  
P Value ◽  
Plasma Samples ◽  
Breast Cancer Patients ◽  
Clinical Value ◽  
Molecular Phenotypes

Purpose: The aim of this study is to identify differential metabolomic signatures in plasma samples of distinct subtypes of breast cancer patients that could be used in clinical practice as diagnostic biomarkers for these molecular phenotypes and to provide a more individualized and accurate therapeutic procedure. Methods: Untargeted LC-HRMS metabolomics approach in positive and negative electrospray ionization mode was used to analyze plasma samples from LA, LB, HER2+ and TN breast cancer patients and healthy controls in order to determine specific metabolomic profiles through univariate and multivariate statistical data analysis. Results: We tentatively identified altered metabolites displaying concentration variations among the four breast cancer molecular subtypes. We found a biomarker panel of 5 candidates in LA, 7 in LB, 5 in HER2 and 3 in TN that were able to discriminate each breast cancer subtype with a false discovery range corrected p-value < 0.05 and a fold-change cutoff value > 1.3. The model clinical value was evaluated with the AUROC, providing diagnostic capacities above 0.85. Conclusion: Our study identifies metabolic profiling differences in molecular phenotypes of breast cancer. This may represent a key step towards therapy improvement in personalized medicine and prioritization of tailored therapeutic intervention strategies.

scholarly journals Lesion-Based Bone Metastasis Detection in Chest Bone Scintigraphy Images of Prostate Cancer Patients Using Pre-Train, Negative Mining, and Deep Learning

Diagnostics ◽  
2021 ◽  
Vol 11 (3) ◽  
pp. 518
Author(s):  
Da-Chuan Cheng ◽  
Te-Chun Hsieh ◽  
Kuo-Yang Yen ◽  
Chia-Hung Kao
Keyword(s):  
Breast Cancer ◽  
Prostate Cancer ◽  
Deep Learning ◽  
Bone Metastasis ◽  
Cancer Patients ◽  
Bone Scintigraphy ◽  
Breast Cancer Patients ◽  
Location Detection ◽  
The Mean ◽  
Mean Sensitivity

This study aimed to explore efficient ways to diagnose bone metastasis early using bone scintigraphy images through negative mining, pre-training, the convolutional neural network, and deep learning. We studied 205 prostate cancer patients and 371 breast cancer patients and used bone scintigraphy data from breast cancer patients to pre-train a YOLO v4 with a false-positive reduction strategy. With the pre-trained model, transferred learning was applied to prostate cancer patients to build a model to detect and identify metastasis locations using bone scintigraphy. Ten-fold cross validation was conducted. The mean sensitivity and precision rates for bone metastasis location detection and classification (lesion-based) in the chests of prostate patients were 0.72 ± 0.04 and 0.90 ± 0.04, respectively. The mean sensitivity and specificity rates for bone metastasis classification (patient-based) in the chests of prostate patients were 0.94 ± 0.09 and 0.92 ± 0.09, respectively. The developed system has the potential to provide pre-diagnostic reports to aid in physicians’ final decisions.

scholarly journals Relationship between platelet-lymphocyte ratio and tumour infiltrating lymphocyte on metastatic in breast cancer patient

2019 ◽  
Vol 7 (6) ◽  
pp. 2045
Author(s):  
Indro Wibowo Sejati ◽  
Ida Bagus Tjakra Wibawa Manuaba ◽  
Putu Anda Tusta ◽  
Gede Budhi Setiawan
Keyword(s):  
Breast Cancer ◽  
Metastatic Breast Cancer ◽  
Cancer Patients ◽  
Cancer Metastasis ◽  
Breast Cancer Subtype ◽  
Metastatic Breast ◽  
Breast Cancer Patients ◽  
Sample Group ◽  
Lymphocyte Ratio ◽  
Platelet Lymphocyte Ratio

Background: Platelet-lymphocyte ratio (PLR) is known associated with the prognosis of distant metastatic breast cancer. Tumor-infiltrating lymphocyte (TIL) in breast cancer also associated with the prognosis of distant metastatic breast cancer. In this study, we will examine the relationship between PLR and TIL, in association with the metastatic incidence in breast cancer.Methods: This research is a retrospective, analytic, cross-sectional study. Data was taken from medical records of breast cancer patients at Sanglah general hospital. Samples were taken by nested sampling by selecting all breast cancer patients from the period of January 1st, 2017, to December 31st, 2018, which had complete medical record data, with total sample 211. The PLR and TIL were calculated and analyzed in relation to metastasis incidence of breast cancer.Results: The sample characteristics were sorted by age, education, occupation, the area of origin, menstrual status, breast cancer staging, breast cancer subtype, TIL levels, lymphovascular invasion (LVI) status, metastatic status, and breast cancer grading. The data were analyzed to know the association of PLR, TIL, confounding factors in relation to metastatic incidences. In the sample group with PLR ≥ 156 10µ /µL, there were 22.9% cases of metastases (p = 0.002). The sample group at low TIL had metastatic event 12.5% with (p=0.442).Conclusions: PLR was associated with higher metastasis in breast cancer patients and low TIL had no association with breast cancer metastasis.

scholarly journals Annotating and Detecting Topics in Social Media Forum and Modelling the Annotation to Derive Directions-A Case Study

2021 ◽  
Author(s):  
Athira B ◽  
Josette Jones ◽  
Sumam Mary Idicula ◽  
Anand Kulanthaivel ◽  
Enming Zhang
Keyword(s):  
Breast Cancer ◽  
Social Media ◽  
Deep Learning ◽  
Cancer Patients ◽  
Learning Algorithm ◽  
Treatment Options ◽  
Healthcare Sector ◽  
Final Decision ◽  
Breast Cancer Patients ◽  
Deep Learning Algorithm

Abstract The widespread influence of social media impacts every aspect of life, including the healthcare sector. Although medics and health professionals are the final decision makers, the advice and recommendations obtained from fellow patients are significant. In this context, the present paper explores the topics of discussion posted by breast cancer patients and survivors on online forums. The study examines an online forum, Breastcancer.org, maps the discussion entries to several topics, and proposes a machine learning model based on a classification algorithm to characterize the topics. To explore the topics of breast cancer patients and survivors, approximately 1000 posts are selected and manually labeled with annotations. In contrast, millions of posts are available to build the labels. A semi-supervised learning technique is used to build the labels for the unlabeled data; hence, the large data are classified using a deep learning algorithm. The deep learning algorithm BiLSTM with BERT word embedding technique provided a better f1-score of 79.5%. This method is able to classify the following topics: medication reviews, clinician knowledge, various treatment options, seeking and providing support, diagnostic procedures, financial issues and implications for everyday life. What matters the most for the patients is coping with everyday living as well as seeking and providing emotional and informational support. The approach and findings show the potential of studying social media to provide insight into patients' experiences with cancer like critical health problems.

Profiling receptor tyrosine kinase fusions in Chinese breast cancers.

2021 ◽  
Vol 39 (15_suppl) ◽  
pp. e15092-e15092
Author(s):  
Zhonghua Tao ◽  
Xichun Hu ◽  
Wen-Ming Cao ◽  
Jianxia Liu ◽  
Ting Li ◽  
...  
Keyword(s):  
Breast Cancer ◽  
Tyrosine Kinase ◽  
Cancer Patients ◽  
Tyrosine Kinases ◽  
Breast Cancer Subtype ◽  
Stage Iv ◽  
Tyrosine Kinase Domain ◽  
Breast Cancers ◽  
Breast Cancer Patients ◽  
Her2 Positive

e15092 Background: Receptor tyrosine kinases (RTKs) are a class of tyrosine kinases that regulate cell-to-cell communication and control a variety of complex biological functions. Dysregulation of RTK signaling partly due to chromosomal rearrangements leads to novel tyrosine kinase fusion oncoproteins which are possibly driver alterations to cancers. Targeting some RTK fusions with specific tyrosine kinases inhibitors (TKIs) is an effective therapeutic strategy across a spectrum of RTK fusion-related cancers. However, there is still a paucity of extensive RTK fusion investigations in breast cancer. We aimed to characterize RTK fusions in Chinese breast cancer patients. Methods: An in-house sequencing database of 1440 Chinese breast cancer patients using a 520-gene NGS sequencing panel was thoroughly reviewed. RTK fusion was defined as an in-frame fusion with the tyrosine kinase domain of the RTK completely retained with the only exception of ERBB2 fusion which was not counted due to its unclear significance. Concomitant mutations and TMB were also analyzed and calculated. Patients’ clinical characteristics were retrieved from case records. Results: 27 RTK fusion-positive breast cancers (12 tissues + 15 plasmas) were identified, patients had a median age of 52 years. Triple-negative breast cancer subtype comprised 37% with luminal and HER2 positive subtypes being 40.8% and 22.2%, respectively. 77.8% of patients were at stage IV and 22.2% at stage I-III. Ten were treatment naïve. RTK fusions occurred in 2% of breast cancers in our database, compared with the prevalence of 0.6% and 1.3% in MSKCC and TCGA, respectively. In the subset of stage IV patients, our database showed a significantly higher RTK fusion frequency than that in MSKCC (3.2% vs. 0.6%, p = 0.013). FGFR2 fusions were seen most commonly (n = 7), followed by RET (n = 4), ROS1 (n = 3), NTRK3 (n = 3), BRAF (n = 2), and NTRK1 (n = 2). Other RTK fusions including ALK, EGFR, FGFR1, FGFR3, MET, and NTRK2 were identified in one patient each. Of note, the normalized abundance of RTK fusion (fusion AF/max AF) correlated negatively with TMB (r = -0.48, p = 0.017). Patients with TMB < 4 (Muts/Mb) displayed a higher fusion abundance than those with TMB ≥ 4 (Muts/Mb) (p = 0.018), suggesting a higher likelihood of subclonal nature for RTK fusions in TMB-high patients. Moreover, CREBBP mutation only co-occurred with FGFR2 fusion (p = 0.012), while NTRK3 fusion and TP53 mutation were mutually exclusive (p = 0.019). Conclusions: This is the first study comprehensively delineating the prevalence and spectrum of RTK fusions in Chinese breast cancers. Further study is ongoing to identify the enriched subpopulation which may benefit from RTK fusion inhibitors.

The glycobiological landscape of breast cancer: insights into galectins and O-GlcNAc homeostasis-related genes

2021 ◽  
Author(s):  
Rada Tazhitdinova ◽  
Alexander V Timoshenko
Keyword(s):  
Breast Cancer ◽  
Cancer Biology ◽  
Breast Cancer Subtype ◽  
The Cancer Genome Atlas ◽  
Correlation Relationship ◽  
Cancer Data ◽  
Normal Tissues ◽  
Genes Encoding ◽  
Genetic Landscape ◽  
Cancer Tissues

Abstract Purpose This study aimed to assess the functional associations between genes of the glycobiological landscape encoding galectins and O-GlcNAc cycle enzymes in the context of breast cancer biology and clinical applications. Methods An in silico analysis of the breast cancer data from The Cancer Genome Atlas was conducted comparing expression, pairwise correlations, and prognostic value for 17 genes encoding galectins, O-GlcNAc cycle enzymes, and cell stemness-related transcription factors. Results Multiple general and breast cancer subtype-specific differences in galectin/O-GlcNAc genetic landscape markers were observed and classified. Specifically, LGALS12 was found to be significantly downregulated in breast cancer tissues across all subtypes while LGALS2 and GFPT1 showed potential as prognostic markers. Remarkably, there was an overall loss of both correlation strength and correlation relationship between expression of galectin/O-GlcNAc landscape genes in the breast cancer samples versus normal tissues. Six gene pairs (GFPT1/LGALS1, GFPT1/LGALS3, GFPT1/LGALS12, GFPT1/KLF4, OGT/LGALS12, and OGT/KLF4) were found to be potential diagnostic markers for breast cancer. Conclusions These findings indicate that the glycobiological landscape of breast cancer underwent significant remodeling, which might be associated with switching galectin gene regulation within a framework of O-GlcNAc homeostasis.

Sign in / Sign up

Export Citation Format

Share Document