scholarly journals Survival models for right censored breast cancer data: theory, application and comparison

F1000Research ◽  
2021 ◽  
Vol 10 ◽  
pp. 1042
Author(s):  
Madiha Liaqat ◽  
Shahid Kamal ◽  
Florian Fischer ◽  
Waqas Fazil
Keyword(s):  
Breast Cancer ◽  
Information Criteria ◽  
Additive Models ◽  
Parametric Models ◽  
Model Complexity ◽  
Parametric Modelling ◽  
Time To Failure ◽  
Cancer Data ◽  
Piecewise Exponential ◽  
Failure Models

Background: Censoring frequently occurs in disease data analysis, which is a key characteristic of time to failure modeling. Typically, time to failure studies are conducted through non-parametric and semi-parametric modelling techniques. Parametric models provide more efficient estimates, but are seldomly used, because of some of the limitations and assumptions which need to be fulfilled to apply them. The aim of this study is to illustrate the theoretical and application limitations and performance of different flexible and standard parametric models to evaluate the prognostic value for mortality risk of breast cancer after recurrence among women. Methods: This article describes the theoretical properties of flexible parametric models and compares their performances to standard parametric models, by studying mortality in women diagnosed with breast cancer. We describe how time to failure data may be analyzed with nonlinear flexible models. In this regard, we apply fractional polynomials, spline models, piecewise exponential models, and piecewise exponential additive mixed models. We also illustrate properties of standard parametric models. All analyses have been conducted with multiple covariates to identify significant predictors. Information criteria have been used to evaluate performances of models. Results: Fractional polynomial and spline-based generalized additive models work well in capturing local fluctuations. Parameter estimation with a piecewise exponential additive mixed model (PAMM) as an extension of the piecewise exponential modelling (PEM) approach automatically penalizes model complexity, which is very helpful to avoid over fitting. Conclusions: Flexible parametric time to failure models are more efficient than standard parametric time to failure models. By incorporating time dependent covariates, PAMM is a good approach to perform in-depth studies of predictors over different finite intervals of follow-up time. Until now, this approach is rarely used in time to failure right censored studies.

Stage migration in breast cancer: Better detection or semantics in staging?

2021 ◽  
Vol 39 (15_suppl) ◽  
pp. e18800-e18800
Author(s):  
Leah Elson ◽  
Nadeem Bilani ◽  
Hong Liang ◽  
Elizabeth Blessing Elimimian ◽  
Diana Saravia ◽  
...  
Keyword(s):  
Breast Cancer ◽  
Cancer Incidence ◽  
Late Stage ◽  
Early Stage ◽  
Breast Cancer Incidence ◽  
Incidence Rates ◽  
Information Criteria ◽  
Stage Migration ◽  
Cancer Data ◽  
Best Fit

e18800 Background: As oncology treatment has evolved to become more individualized, prognostic rationale has also undergone important changes. In breast cancer, disease staging was historically based upon anatomic features of the primary tumor, in combination with involvement of adjacent/distant tissues. However, as the understanding of molecular/genomic involvement became more advanced, staging definitions were redefined to incorporate receptors, histologic grade, and genetic expression. In this analysis, we use autoregressive integrated moving average (ARIMA) forecasting to understand how AJCC updates to prognostic definitions have contributed to stage migration, and to comment on whether better detection, or definitional changes, may be responsible for the increasing incidence in early stage breast cancer. Methods: In this time series forecast, ARIMA models, per stage (early: stage I/II vs. late: stage III/IV) were constructed based on 2004-2016 historic breast cancer incidence rates, as reported by the NCDB. Multiple models were generated, using differing autoregressive parameters; the most predictive model was chosen using the lowest Bayesian Information Criteria (BIC), and mean absolute percentage error (MAPE) to ensure best fit. Similar methodology has already been published to predict prostate cancer incidence. The best fit models were applied to forecast annual incidence, in the NCDB, in 2017. These data were compared to the real-world data captured in 2017. Statistics were performed using modeling systems in SPSS, version 27. Results: n=1,661,971 cases were included for these models, and 12 years of pre-AJCC updated NCDB breast cancer data were used. Using ARIMA modeling, best fit, stationary averages were identified, with autoregressive and difference terms which contributed to the lowest BIC, and MAPE < 5%, for both models. The best fit models forecasted 2017 incidence, by stage, without AJCC updates to staging criteria, and this data is compared to actual 2017 incidence with current updated AJCC 8th staging criteria (Table). Conclusions: During 2017, the first year of AJCC staging updates, there was an observed decrease in late stage diagnoses, and increase in early stage diagnoses, when compared with incidence rates that were forecasted using the old, anatomic AJCC criteria. Therefore, part of the stage migration noted may be a product of staging semantics, using updated definitions. Confirming appropriate improvement in long-term outcomes, based on new staging would be helpful. It is also important for clinicians and public health officials to bear this in-mind when interpreting epidemiologic data, for allocating resources, as shifts in staging may be a product of guideline changes, and not necessarily screening efficacy or early detection only.[Table: see text]

scholarly journals Introducing Hurdle_T as a Count Regression Model with High Potency in Solving Problems

2020 ◽  
pp. 37-50
Author(s):  
Ojekudo, Nathaniel Akpofure ◽  
Akpan, Nsikan Paul
Keyword(s):  
Breast Cancer ◽  
Regression Model ◽  
Male Breast Cancer ◽  
Negative Binomial ◽  
Health Sector ◽  
Information Criteria ◽  
Male Breast ◽  
Data Sets ◽  
Cancer Data ◽  
Binary Model

Count data regression models exhibit different strengths and weaknesses in their bids to solving problems. The study considers six count models namely Poisson Regression Model (PRM), Negative Regression Model (NBRM), Zero Inflated Poisson (ZIP), Zero Inflated Negative Binomial (ZINB), Zero Truncated Poisson (ZTP) and Zero Truncated Negative Binomial (ZTNB) and an additional model called hurdle_T. These models are used to analyze two health data sets. The data on male breast cancer reveals that male breast cancer cuts across all age brackets or categories but it is more prevalent between the ages of 50 and 60. The PRM yields a better result than the NBRM in the case of cancer data as shown by the information criteria. The analysis of the second data, which is on doctor’s visit reveals that ZINB yields a better result than the other five models, followed by NBRM, then the ZTNB before their Poisson counterparts. The hurdle_T model shows the propensity of each coefficient as reflected by the positive count in the Tobit (Binary) model. The study also shows that at 65 years and above, gender has significant effect on doctor’s visit. In particular, females, more than males attract more doctors’ visit in the said age range. Government policies should provide more funds in the health sector to accommodate cancer cases in terms of the provision of awareness, studies/ research and infrastructural development. Males should be encouraged to visit clinics especially in their late forties and above for breast cancer related checkup. At age 65 and above, doctors visit to patients are frequent, especially to females. Policy of government in the health sector should accommodate a favourable adjustment in the budget to take care of doctors’ visit.

Bayesian Frailty Model for Time to Event Breast Cancer Data

2011 ◽  
Vol 4 (2) ◽  
pp. 8-12
Author(s):  
Leo Alexander T Leo Alexander T ◽  
◽  
Pari Dayal L Pari Dayal L ◽  
Valarmathi S Valarmathi S ◽  
Ponnuraja C Ponnuraja C ◽  
...  
Keyword(s):  
Breast Cancer ◽  
Frailty Model ◽  
Breast Cancer Data ◽  
Time To Event ◽  
Cancer Data

Perbandingan CART dan Random Forest untuk Deteksi Kanker berbasis Klasifikasi Data Microarray

2020 ◽  
Vol 4 (5) ◽  
pp. 805-812
Author(s):  
Riska Chairunisa ◽  
Adiwijaya ◽  
Widi Astuti
Keyword(s):  
Breast Cancer ◽  
Lung Cancer ◽  
Ovarian Cancer ◽  
Random Forest ◽  
Classification And Regression Tree ◽  
Classification Method ◽  
Discrete Wavelet ◽  
Prostate Tumors ◽  
Cancer Data ◽  
Colon Tumors

Cancer is one of the deadliest diseases in the world with a mortality rate of 57,3% in 2018 in Asia. Therefore, early diagnosis is needed to avoid an increase in mortality caused by cancer. As machine learning develops, cancer gene data can be processed using microarrays for early detection of cancer outbreaks. But the problem that microarray has is the number of attributes that are so numerous that it is necessary to do dimensional reduction. To overcome these problems, this study used dimensions reduction Discrete Wavelet Transform (DWT) with Classification and Regression Tree (CART) and Random Forest (RF) as classification method. The purpose of using these two classification methods is to find out which classification method produces the best performance when combined with the DWT dimension reduction. This research use five microarray data, namely Colon Tumors, Breast Cancer, Lung Cancer, Prostate Tumors and Ovarian Cancer from Kent-Ridge Biomedical Dataset. The best accuracy obtained in this study for breast cancer data were 76,92% with CART-DWT, Colon Tumors 90,1% with RF-DWT, lung cancer 100% with RF-DWT, prostate tumors 95,49% with RF-DWT, and ovarian cancer 100% with RF-DWT. From these results it can be concluded that RF-DWT is better than CART-DWT.  

AGE FEATURES OF MOLECULAR-BIOLOGICAL SUBTYPES OF BREAST CANCER IN THE REPUBLIC OF UZBEKISTAN

2018 ◽  
Vol 64 (2) ◽  
pp. 196-199
Author(s):  
Gulya Miryusupova ◽  
G. Khakimov ◽  
N. Shayusupov
Keyword(s):  
Breast Cancer ◽  
Cancer Care ◽  
Indigenous Women ◽  
Morbidity And Mortality ◽  
Breast Cancer Data ◽  
Female Population ◽  
Cancer Data ◽  
Age Related ◽  
The Republic ◽  
Age Features

According to the results of breast cancer data in the Republic of Uzbekistan in addition to the increase in morbidity and mortality from breast cancer among women the presence of age specific features among indigenous women in the direction of “rejuvenating” of the disease with all molecular-biological (phenotypic) subtypes of breast cancer were marked. Within the framework of age-related features the prevalence of the least favorable phenotypes of breast cancer was found among indigenous women: Her2/neu hyperexpressive and three times negative subtype of breast cancer. The data obtained made it possible to build a so-called population “portrait” of breast cancer on the territory of the Republic, which in turn would contribute to further improvement of cancer care for the female population of the country.

scholarly journals Investigation of monogenic causes of familial breast cancer: data from the BEACCON case-control study

2021 ◽  
Vol 7 (1) ◽  
Author(s):  
Na Li ◽  
Belle W. X. Lim ◽  
Ella R. Thompson ◽  
Simone McInerny ◽  
Magnus Zethoven ◽  
...  
Keyword(s):  
Breast Cancer ◽  
High Risk ◽  
Candidate Genes ◽  
Familial Aggregation ◽  
Loss Of Function ◽  
Cancer Data ◽  
Pathogenic Variants ◽  
New Genes ◽  
Predisposing Genes ◽  
Predisposition Genes

AbstractBreast cancer (BC) has a significant heritable component but the genetic contribution remains unresolved in the majority of high-risk BC families. This study aims to investigate the monogenic causes underlying the familial aggregation of BC beyond BRCA1 and BRCA2, including the identification of new predisposing genes. A total of 11,511 non-BRCA familial BC cases and population-matched cancer-free female controls in the BEACCON study were investigated in two sequencing phases: 1303 candidate genes in up to 3892 cases and controls, followed by validation of 145 shortlisted genes in an additional 7619 subjects. The coding regions and exon–intron boundaries of all candidate genes and 14 previously proposed BC genes were sequenced using custom designed sequencing panels. Pedigree and pathology data were analysed to identify genotype-specific associations. The contribution of ATM, PALB2 and CHEK2 to BC predisposition was confirmed, but not RAD50 and NBN. An overall excess of loss-of-function (LoF) (OR 1.27, p = 9.05 × 10−9) and missense (OR 1.27, p = 3.96 × 10−73) variants was observed in the cases for the 145 candidate genes. Leading candidates harbored LoF variants with observed ORs of 2–4 and individually accounted for no more than 0.79% of the cases. New genes proposed by this study include NTHL1, WRN, PARP2, CTH and CDK9. The new candidate BC predisposition genes identified in BEACCON indicate that much of the remaining genetic causes of high-risk BC families are due to genes in which pathogenic variants are both very rare and convey only low to moderate risk.

scholarly journals The INTREST registry: protocol of a multicenter prospective cohort study of predictors of women’s response to integrative breast cancer treatment

BMC Cancer ◽  
2021 ◽  
Vol 21 (1) ◽  
Author(s):  
Heidemarie Haller ◽  
Petra Voiß ◽  
Holger Cramer ◽  
Anna Paul ◽  
Mattea Reinisch ◽  
...  
Keyword(s):  
Breast Cancer ◽  
Clinical Trial ◽  
Cohort Study ◽  
Cancer Treatment ◽  
Prospective Cohort Study ◽  
Cancer Care ◽  
Prospective Cohort ◽  
Breast Cancer Treatment ◽  
Cancer Data ◽  
Patient Reported

Abstract Background Cancer registries usually assess data of conventional treatments and/or patient survival. Beyond that, little is known about the influence of other predictors of treatment response related to the use of complementary therapies (CM) and lifestyle factors affecting patients’ quality and quantity of life. Methods INTREST is a prospective cohort study collecting register data at multiple German certified cancer centers, which provide individualized, integrative, in- and outpatient breast cancer care. Patient-reported outcomes and clinical cancer data of anticipated N = 715 women with pTNM stage I-III breast cancer are collected using standardized case report forms at the time of diagnosis, after completing neo−/adjuvant chemotherapy, after completing adjuvant therapy (with the exception of endocrine therapy) as well as 1, 2, 5, and 10 years after baseline. Endpoints for multivariable prediction models are quality of life, fatigue, treatment adherence, and progression-based outcomes/survival. Predictors include the study center, sociodemographic characteristics, histologic cancer and comorbidity data, performance status, stress perception, depression, anxiety, sleep quality, spirituality, social support, physical activity, diet behavior, type of conventional treatments, use of and belief in CM treatments, and participation in a clinical trial. Safety is recorded following the Common Terminology Criteria for Adverse Events. Discussion This trial is currently recruiting participants. Future analyses will allow to identify predictors of short- and long-term response to integrative breast cancer treatment in women, which, in turn, may improve cancer care as well as quality and quantity of life with cancer. Trial registration German Clinical Trial Register DRKS00014852. Retrospectively registered at July 4th, 2018.

Sub-Ethnicity and Survivorship Experience: Asian American Breast Cancer Survivors

2021 ◽  
pp. 019394592110319
Author(s):  
Wonshik Chee ◽  
Eun-Ok Im
Keyword(s):  
Breast Cancer ◽  
Cancer Survivors ◽  
Asian American ◽  
Breast Cancer Survivors ◽  
Short Form ◽  
Secondary Analysis ◽  
Future Research ◽  
Japanese American ◽  
Personal Resource ◽  
Cancer Data

The purpose of the study was to explore the associations of sub-ethnicity to the survivorship experience of Asian American breast cancer survivors and identify the multiple factors that influenced their survivorship experience. This was a secondary analysis of the data among 94 Asian American breast cancer survivors from a larger ongoing study. Instruments included: questions on background characteristics, the perceived isolation scale, the Personal Resource Questionnaire, the Memorial Symptom Assessment Scale-Short Form, and the Functional Assessment of Cancer Therapy-Breast Cancer. Data were analyzed using hierarchical logistic and multiple regression analyses. After controlling for other factors, being a Japanese American (ref. = being a Chinese American) was significantly associated with pain scores (odds ratio [OR] = −0.32, p < .01), symptom distress scores ( β = −0.27, p < .01), and the quality of life scores ( β = 0.22, p = .03). Sub-ethnic variations in cultural attitudes, values, and beliefs need to be considered in future research/practice with Asian American breast cancer survivors.

Breast cancer diagnosis using multiple activation deep neural network

2021 ◽  
pp. 1063293X2110251
Author(s):  
K Vijayakumar ◽  
Vinod J Kadam ◽  
Sudhir Kumar Sharma
Keyword(s):  
Breast Cancer ◽  
Neural Network ◽  
Deep Neural Network ◽  
Breast Cancer Diagnosis ◽  
Activation Function ◽  
Linear Functions ◽  
Cancer Data ◽  
Final Layer ◽  
Improved Performance ◽  
Hidden Layer

Deep Neural Network (DNN) stands for multilayered Neural Network (NN) that is capable of progressively learn the more abstract and composite representations of the raw features of the input data received, with no need for any feature engineering. They are advanced NNs having repetitious hidden layers between the initial input and the final layer. The working principle of such a standard deep classifier is based on a hierarchy formed by the composition of linear functions and a defined nonlinear Activation Function (AF). It remains uncertain (not clear) how the DNN classifier can function so well. But it is clear from many studies that within DNN, the AF choice has a notable impact on the kinetics of training and the success of tasks. In the past few years, different AFs have been formulated. The choice of AF is still an area of active study. Hence, in this study, a novel deep Feed forward NN model with four AFs has been proposed for breast cancer classification: hidden layer 1: Swish, hidden layer, 2:-LeakyReLU, hidden layer 3: ReLU, and final output layer: naturally Sigmoidal. The purpose of the study is twofold. Firstly, this study is a step toward a more profound understanding of DNN with layer-wise different AFs. Secondly, research is also aimed to explore better DNN-based systems to build predictive models for breast cancer data with improved accuracy. Therefore, the benchmark UCI dataset WDBC was used for the validation of the framework and evaluated using a ten-fold CV method and various performance indicators. Multiple simulations and outcomes of the experimentations have shown that the proposed solution performs in a better way than the Sigmoid, ReLU, and LeakyReLU and Swish activation DNN in terms of different parameters. This analysis contributes to producing an expert and precise clinical dataset classification method for breast cancer. Furthermore, the model also achieved improved performance compared to many established state-of-the-art algorithms/models.

Sign in / Sign up

Export Citation Format

Share Document