BIOSTATISTICS AND BIOINFORMATICS

This editorial seeks to paint a clearer picture of Big Data in healthcare and the benefits for better cancer care. It gives a simple, widely used definition of Big Data and explains some of its pertinent characteristics (volume, velocity, variety, veracity) with respect to healthcare. A brief discussion of some solutions provided by Big Data for healthcare challenges together with vivid examples where Big Data have been implemented to improve operational efficiency and clinical excellence for hospitals. Big Data has been gradually changing the face of healthcare, with the promise to improve cancer care, decision making in real-time via decision support systems(DSS), and better operations in healthcare institutions. However, making Big Data Findable Accessible Interoperable and Reusable (FAIR) in healthcare would be an essential improvement not only to circumvent the data-sharing problem but also to open up the possibility to use and reuse healthcare big data and still draw reliable conclusions.

Pre-requisites to better understand the chapter: knowledge of the major steps and procedures of developing a clinical prediction model.

Logical position of the chapter with respect to the previous chapter: in the last chapters, you have learned how to develop and validate a clinical prediction model. You have been learning logistic regression as the main algorithm to build the model. However, several different more complex algorithms can be used to build a clinical prediction model. In this chapter, the main machine learning-based algorithms will be presented to you.

Learning objectives: you will be presented with the definitions of machine learning, supervised, and unsupervised learning. The major algorithms for the last two categories will be introduced.

While the 10-year survival rate for localized prostate cancer patients is very good (>98%), side effects of treatment may limit quality of life significantly. Erectile dysfunction (ED) is a common burden associated with increasing age as well as prostate cancer treatment. Although many studies have investigated the factors affecting erectile dysfunction (ED) after prostate cancer treatment, only limited studies have investigated whether ED can be predicted before the start of treatment. The advent of machine learning (ML) based prediction tools in oncology offers a promising approach to improve accuracy of prediction and quality of care. Predicting ED may help aid shared decision making by making the advantages and disadvantages of certain treatments clear, so that a tailored treatment for an individual patient can be chosen. This study aimed to predict ED at 1-year and 2year post-diagnosis based on patient demographics, clinical data and patient-reported outcomes (PROMs) measure...

Pre-requisites to better understand the chapter: knowledge of the major steps and procedures of developing a clinical prediction model.Logical position of the chapter with respect to the previous chapter: in the last chapters, you have learned how to develop and validate a clinical prediction model. You have been learning logistic regression as main algorithm to build the model. However, several different more complex algorithms can be used to build a clinical prediction model. In this chapter, the main machine learning based algorithms will be presented to you.Learning objectives: you will be presented with the definitions of: machine learning, supervised and unsupervised learning. The major algorithms for the last two categories will be introduced.

This study aims to analyze the relationship between the available variables and treatment compliance in elderly cancer patients treated with radiotherapy and to establish a decision tree model to guide caregivers in their decision-making process. For this purpose, 456 patients over 74 years of age who received radiotherapy between 2005 and 2017 were included in this retrospective analysis. The outcome of interest was radiotherapy compliance, determined by whether patients completed their scheduled radiotherapy treatment (compliance means they completed their treatment and noncompliance means they did not). A bootstrap (B = 400) technique was implemented to select the best tuning parameters to establish the decision tree. The developed decision tree uses patient status, the Charlson comorbidity index, the Eastern Cooperative Oncology Group Performance scale, age, sex, cancer type, health insurance status, radiotherapy aim, and fractionation type (conventional fractionation versus hyp...

While the 10-year survival rate for localized prostate cancer patients is very good (>98%), side effects of treatment may limit quality of life significantly. Erectile dysfunction (ED) is a common burden associated with increasing age as well as prostate cancer treatment. Although many studies have investigated the factors affecting erectile dysfunction (ED) after prostate cancer treatment, only limited studies have investigated whether ED can be predicted before the start of treatment. The advent of machine learning (ML) based prediction tools in oncology offers a promising approach to improve the accuracy of prediction and quality of care. Predicting ED may help aid shared decision-making by making the advantages and disadvantages of certain treatments clear, so that a tailored treatment for an individual patient can be chosen. This study aimed to predict ED at 1-year and 2-year post-diagnosis based on patient demographics, clinical data and patient-reported outcomes (PROMs) me...

Introduction: Urinary incontinence (UI) is a common side effect of prostate cancer treatment, but in clinical practice, it is difficult to predict. Machine learning (ML) models have shown promising results in predicting outcomes, yet the lack of transparency in complex models known as "black-box" has made clinicians wary of relying on them in sensitive decisions. Therefore, finding a balance between accuracy and explainability is crucial for the implementation of ML models. The aim of this study was to employ three different ML classifiers to predict the probability of experiencing UI in men with localized prostate cancer 1-year and 2-year after treatment and compare their accuracy and explainability. Methods: We used the ProZIB dataset from the Netherlands Comprehensive Cancer Organization (Integraal Kankercentrum Nederland; IKNL) which contained clinical, demographic, and PROM data of 964 patients from 65 Dutch hospitals. Logistic Regression (LR), Random Forest (RF), and Support Vector Machine (SVM) algorithms were applied to predict (in)continence after prostate cancer treatment. Results: All models have been externally validated according to the TRIPOD Type 3 guidelines and their performance was assessed by accuracy, sensitivity, specificity, and AUC. While all three models demonstrated similar performance, LR showed slightly better accuracy than RF and SVM in predicting the risk of UI one year after prostate cancer treatment, achieving an accuracy of 0.75, a sensitivity of 0.82, and an AUC of 0.79. All models for the 2year outcome performed poorly in the validation set, with an accuracy of 0.6 for LR, 0.65 for RF, and 0.54 for SVM.

Conclusion In cervical cancer patients treated with chemoradiation, LMR over and above established prognostic factors including FIGO Stage, hemoglobin and nutritional status seems to be an important independent prognostic factor for DFS and OAS. Further research to get more insights and clinical relevance is warranted.

Machine learning is an application of Artificial Intelligence (AI). AI refers to the ability to program computers (or more in general machines) that are able to solve complicated and usually very time-consuming tasks [1]. An example of a time consuming and complicated task is the extraction of useful information (data mining) from a large amount of unstructured clinical data ('big data'). 9.3 How Do We Use Machine Learning in Clinical Prediction Modelling? As you learned in previous chapters, after having prepared your data, you develop a clinical prediction model based on available data. In the model, particular properties of your data ('features') will be used to predict your outcome of interest. A particular statistical algorithm is used to learn the 'features' that are most representative and relate them to the predicted outcome. In the previous chapter, only the logistic regression algorithm has been presented to you. However, more complex machine learning-based algorithms exist. These algorithms can be divided into two categories: supervised and unsupervised [2].

People interested in the research are advised to contact the author for the final version of the publication, or visit the DOI to the publisher's website. • The final author version and the galley proof are versions of the publication after peer review. • The final published version features the final layout of the paper including the volume, issue and page numbers. Link to publication General rights Copyright and moral rights for the publications made accessible in the public portal are retained by the authors and/or other copyright owners and it is a condition of accessing publications that users recognise and abide by the legal requirements associated with these rights. • Users may download and print one copy of any publication from the public portal for the purpose of private study or research. • You may not further distribute the material or use it for any profit-making activity or commercial gain • You may freely distribute the URL identifying the publication in the public portal. If the publication is distributed under the terms of Article 25fa of the Dutch Copyright Act, indicated by the "Taverne" license above, please follow below link for the End User Agreement:

Introduction: Urinary incontinence (UI) is a common side effect of prostate cancer treatment, but in clinical practice, it is difficult to predict. Machine learning (ML) models have shown promising results in predicting outcomes, yet the lack of transparency in complex models known as "black-box" has made clinicians wary of relying on them in sensitive decisions. Therefore, finding a balance between accuracy and explainability is crucial for the implementation of ML models. The aim of this study was to employ three different ML classifiers to predict the probability of experiencing UI in men with localized prostate cancer 1-year and 2-year after treatment and compare their accuracy and explainability. Methods: We used the ProZIB dataset from the Netherlands Comprehensive Cancer Organization (Integraal Kankercentrum Nederland; IKNL) which contained clinical, demographic, and PROM data of 964 patients from 65 Dutch hospitals. Logistic Regression (LR), Random Forest (RF), and Support Vector Machine (SVM) algorithms were applied to predict (in)continence after prostate cancer treatment. Results: All models have been externally validated according to the TRIPOD Type 3 guidelines and their performance was assessed by accuracy, sensitivity, specificity, and AUC. While all three models demonstrated similar performance, LR showed slightly better accuracy than RF and SVM in predicting the risk of UI one year after prostate cancer treatment, achieving an accuracy of 0.75, a sensitivity of 0.82, and an AUC of 0.79. All models for the 2year outcome performed poorly in the validation set, with an accuracy of 0.6 for LR, 0.65 for RF, and 0.54 for SVM.

While the 10-year survival rate for localized prostate cancer patients is very good (>98%), side effects of treatment may limit quality of life significantly. Erectile dysfunction (ED) is a common burden associated with increasing age as well as prostate cancer treatment. Although many studies have investigated the factors affecting erectile dysfunction (ED) after prostate cancer treatment, only limited studies have investigated whether ED can be predicted before the start of treatment. The advent of machine learning (ML) based prediction tools in oncology offers a promising approach to improve the accuracy of prediction and quality of care. Predicting ED may help aid shared decision-making by making the advantages and disadvantages of certain treatments clear, so that a tailored treatment for an individual patient can be chosen. This study aimed to predict ED at 1-year and 2-year postdiagnosis based on patient demographics, clinical data and patient-reported outcomes (PROMs) measured at diagnosis. We used a subset of the ProZIB dataset collected by the Netherlands Comprehensive Cancer Organization (Integraal Kankercentrum Nederland; IKNL) that contained information on 964 localized prostate cancer cases from 69 Dutch hospitals for model training and external validation. Two models were generated using a logistic regression algorithm coupled with Recursive Feature Elimination (RFE). The first predicted ED 1 year post-diagnosis and required 10 pre-treatment variables; the second predicted ED 2 years post-diagnosis with 9 pre-treatment variables. The validation AUCs were 0.84 and 0.81 for 1 year and 2 years post-diagnosis respectively. To immediately allow patients and clinicians to use these models in the clinical decision-making process, nomograms were

Background: The incidence of lung cancer has been increasing. Healthcare providers are trying to acquire more knowledge of the disease's biology to treat their patients better. However, the information available is more than humans can comprehend. Predictive models such as Bayesian networks, which can intricately represent causal relations between variables, are suitable structures to model this information. However, the conventional methods for developing Bayesian network structures are limited in healthcare. abstracts Annals of Oncology Volume 33-Issue S2-2022 S107

Objective: HIV and AIDS have adverse effects on the livelihood of adolescents especially when parents die or are poor due to ill health. Mildmay Uganda has an apprenticeship programme that helps such adolescents acquire skills in order to improve their quality of life. The aim of this study was to examine the role of apprenticeship in improving the livelihood of HIV-positive adolescents who prematurely dropped out of school. Methods: A retrospective review of data of all adolescents who had been enrolled on the apprenticeship programme from January 2012 to January 2014 was done to document their progress and any lessons learnt. A vulnerability index tool was used to assess vulnerability levels and at same time to measure the progress and improvement of livelihood. Results: A total of twenty (20) adolescents were enrolled on the apprenticeship programme during the stated period. Sixty percent were male and were in the age range of 17-24 years. Sixteen out of the 20 (80%) were able to use the acquired skills through apprenticeship to improve their livelihood. The other four adolescents completed apprenticeship, but did not practice their acquired skills. One male opted to venture into another business and three females abandoned their vocations when they got married. The vulnerability level assessment outcome showed a graduation from extremely poor to poor. Adolescents were able to meet their basic needs, keep clinic appointments and were emotionally stable by end of period of review. Conclusion: Apprenticeship is one way that can be used successfully to improve the livelihood of adolescents who prematurely drop out of school.

Background: We aimed to assess the relationship between gender and survival among adult patients newly enrolled on antiretroviral therapy (ART) in Uganda. We also specifically examined the role of antenatal services in favoring women's access to HIV care. Methods: From an observational cohort study, we assessed survival and used logistic regression and differences in means to compare men and women who did not access care through antenatal services. Differences were assessed on measures of disease progression (WHO stage and CD4 count) and demographic (age, marital status, and education), behavioral (sexual activity, disclosure to partner, and testing), and clinical variables (hepatitis B and C, syphilis, malaria, and anemia). A mediational analysis that considered gender as the initial variable, time to death as the outcome, initial CD4 count as the mediator, and age as a covariate was performed using an accelerated failure time model with a Weibull distribution. Results: Between 2004 and 2011, a total of 4775 patients initiated ART, and after exclusions 4537 (93.2%) were included in analysis. Men initiating ART were more likely to have a WHO disease stage III or IV (odds ratio: 1.46, 95% confidence interval [CI]: 1.29-1.66), and lower CD4 cell counts compared to women (median baseline CD4 124 cells/mm 3 , interquartile range [IQR]: 43-205 versus 147 cells/mm 3 , IQR: 68-212, P-value , 0.0001). Men were at an increased risk of death compared to women (hazard ratio: 1.38, 95% CI: 1.03-1.83). Baseline CD4 cell counts accounted for 43% of the increased risk of death in men (95% CI: 22%-113%). Access to care via antenatal services did not explain differences in outcomes. Conclusion: In this cohort there is a marked increase in risk of mortality for men and approximately half of it can be attributed to their later engagement in care. More effort is required to engage men in care in a timely manner.

Background: HIV-infected patients are prone to nutritional problems and are at risk to inflammation-driven premature endothelial injury. Elevated homocysteine (Hc) levels are associated with an increased risk for cardiovascular events. Deficiency of the vitamins B6, B12 and folic acid may contribute to endothelial as well as neurocognitive and/or neuropathic complications. MethodsM p=0.04 [Chi2 test]). No other associations were detectable. Conclusion: In our study, hyperhomocysteinemia is not related to a distinct vitamin deficiency and is independently associated with a significant risk of having AIDS.The elevation may rather due to inflammatory events than to metabolic deficiencies. If vitamin supplementation ameliorates hyperhomocysteinemia and its potential consequences has to be proven. The significance of vitamin deficiency itself in HIV-infected individuals remains to be determined.

A high prevalence of epilepsy has been observed in onchocerciasis endemic areas in the Democratic Republic of Congo (DRC). With this study we aimed to investigate whether Onchocerca volvulus infection is a risk factor for developing epilepsy in onchocerciasis endemic regions in the DRC. Between October and December 2015, a multi-centre case control study was performed in onchocerciasis endemic health zones (HZ) in the DRC: one study site was situated in Tshopo Province in the HZ of Wanierukula (village of Salambongo) where there had been 13 annual community distributions of treatment with ivermectin (CDTI), a second was situated in Ituri Province in the HZ of Logo (village of Draju) where ivermectin had never been distributed and in the HZ of Rethy (village of Rassia) where there had been THREE CDTI annual campaigns before the study. Individuals with unprovoked convulsive epilepsy of unknown etiology were enrolled as cases (n = 175). Randomly selected healthy members of families wit...

The survey also revealed a prevalence of 5.5% for HIV. Given the wide geographic overlap in occurrence of these diseases, co-infection would be expected to have a significant impact on the health of co-infected persons. Earlier studies have failed to reveal important interactions between the two diseases (