Prediction of cancer incidence rates for the European continent using machine learning models

Toward a comprehensive life-cycle carcinogenic impact assessment: A statistical regression approach based on cancer burden

The current approach to life cycle carcinogenic impact assessment (LCCA) is hindered by its static and linear characteristics. This situation prevents the accurate prediction of the incidence, associated damage, and potential economic burden of cancer. This study explores a highly comprehensive pathway for LCCA assessment. It uses the impacts of Tracheal, bronchus, and lung (TBL) predicted by the LCCA of China's coal power industry through a screened statistical regression model as the research target. The latest global burden of disease estimates is utilized to quantify the health damage from TBL incidence, whereas an approach combining the actual cost of health and human capital is applied to further assess the economic burden of TBL. Findings indicate that the traditional and static LCCA method, which relies on animal toxicity data, can lead to underestimations in actual LCCA. The interaction among spatiotemporal meteorological factors, epidemiological cancer disease burden, and socioeconomic behaviors allows exhibits nonlinearity due to the changes in the combined toxicity of mixed key substances. Following the active implementation of ultralow emission and energy-saving transformations in China's coal power industry, the national percentage of TBL cancer incidence caused by pollutants from the coal power industry decreased from 25.2 % in 2004 to 11.5 % in 2020. Results indicate that the established dynamic LCCA model based on temporal and spatial climate, socioeconomic, and epidemiological cancer data can be feasibly employed for the accurate impact evaluation and mitigation of carcinogens in practical applications.

Decrease in the number of new cancer diagnoses during the first year of the COVID-19 pandemic - cohort study of 3.5 million individuals in western Poland

Introduction

The COVID-19 pandemic has considerably affected healthcare systems worldwide and is expected to influence cancer incidence, mortality, stage at diagnosis, and survival. This study aimed to assess COVID-19-related changes in cancer incidence observed in 2020 in the Greater Poland region.

Materials and methods

Data from the Greater Poland Cancer Registry on cancer patients diagnosed between 2010 and 2020 were analysed. To quantify the change in the number of incident cancer cases during the COVID-19 pandemic, we calculated the standardized incidence ratio (SIR) and the incidence rate difference (IRD) to assume the pandemic-attributable gap in cancer incidence.

Results

In 2020, in Greater Poland, the expected number of new cancer cases was 18 154 (9 226 among males and 8 927 among females), while the observed number was 14 770 (7 336 among males and 7 434 among females). The registered number of cancer cases decreased in 2020 by 20% (SIR 0·80, 95% CI 0·78 to 0·81) and 17% (SIR 0·83, 95% CI 0·81 to 0·85) in males and females, respectively. Among men, the most significant difference was reported for myeloma (SIR 0·59, 95% CI 0·45 to 0·77), among women for bone cancer (SIR 0·47, 95% CI 0·20 to 0·93). In females the observed incidence was higher than expected for cancer of an unspecified site (SIR 1·19, 95% CI 1·01 to 1·38). In our study, the decrease in new cancer cases was greater in males than in females.

Discussion

The observed incidence was affected in most cancer sites, with the most significant deviation from the expected number in the case of myeloma. An increase in the observed incidence was reported only in women diagnosed with cancer of an unspecified site, which might reflect shortages in access to oncological diagnostics.

Exploring the association of PM2.5 with lung cancer incidence under different climate zones and socioeconomic conditions from 2006 to 2016 in China

Air pollution generated by urbanization and industrialization poses a significant negative impact on public health. Particularly, fine particulate matter (PM2.5) has become one of the leading causes of lung cancer mortality worldwide. The relationship between air pollutants and lung cancer has aroused global widespread concerns. Currently, the spatial agglomeration dynamic of lung cancer incidence (LCI) has been seldom discussed, and the spatial heterogeneity of lung cancer's influential factors has been ignored. Moreover, it is still unclear whether different socioeconomic levels and climate zones exhibit modification effects on the relationship between PM2.5 and LCI. In the present work, spatial autocorrelation was adopted to reveal the spatial aggregation dynamic of LCI, the emerging hot spot analysis was introduced to indicate the hot spot changes of LCI, and the geographically and temporally weighted regression (GTWR) model was used to determine the affecting factors of LCI and their spatial heterogeneity. Then, the modification effects of PM2.5 on the LCI under different socioeconomic levels and climatic zones were explored. Some findings were obtained. The LCI demonstrated a significant spatial autocorrelation, and the hot spots of LCI were mainly concentrated in eastern China. The affecting factors of LCI revealed an obvious spatial heterogeneity. PM2.5 concentration, nighttime light data, 2 m temperature, and 10 m u-component of wind represented significant positive effects on LCI, while education-related POI exhibited significant negative effects on LCI. The LCI in areas with low urbanization rates, low education levels, and extreme climate conditions was more easily affected by PM2.5 than in other areas. The results can provide a scientific basis for the prevention and control of lung cancer and related epidemics.

New Regional Dynamic Cancer Model across the European Union

BACKGROUND

Can increasing levels of economic wealth significantly influence changes in cancer incidence and mortality rates?

METHODS

We investigated this issue by means of regression analyses based on the study of incidence and mortality indicators for lip, oral cavity, and pharyngeal; colon; pancreatic; lung; leukaemia; brain and central nervous system cancers in correlation with the levels of economic welfare and financial allocations to health at the level of the European Union member states, with the exception of Luxembourg and Cyprus for which there are no official statistical data reported.

RESULTS

The results of the study showed that there were significant disparities both regionally and by gender, requiring corrective public policy measures that were formulated in this study.

CONCLUSIONS

The conclusions highlight the main findings of the study in terms of the evolution of the disease, present the significant aspects that characterise the evolution of each type of cancer during the period analysed (1993-2021), and highlight the novelty and limitations of the study and future directions of research. As a result, increasing economic welfare is a potential factor in halting the effects of cancer incidence and mortality at the population level, while the financial allocations to health of EU member countries' budgets are a drawback due to large regional disparities.

The prediction of distant metastasis risk for male breast cancer patients based on an interpretable machine learning model

OBJECTIVES

This research was designed to compare the ability of different machine learning (ML) models and nomogram to predict distant metastasis in male breast cancer (MBC) patients and to interpret the optimal ML model by SHapley Additive exPlanations (SHAP) framework.

METHODS

Four powerful ML models were developed using data from male breast cancer (MBC) patients in the SEER database between 2010 and 2015 and MBC patients from our hospital between 2010 and 2020. The area under curve (AUC) and Brier score were used to assess the capacity of different models. The Delong test was applied to compare the performance of the models. Univariable and multivariable analysis were conducted using logistic regression.

RESULTS

Of 2351 patients were analyzed; 168 (7.1%) had distant metastasis (M1); 117 (5.0%) had bone metastasis, and 71 (3.0%) had lung metastasis. The median age at diagnosis is 68.0 years old. Most patients did not receive radiotherapy (1723, 73.3%) or chemotherapy (1447, 61.5%). The XGB model was the best ML model for predicting M1 in MBC patients. It showed the largest AUC value in the tenfold cross validation (AUC:0.884; SD:0.02), training (AUC:0.907; 95% CI: 0.899-0.917), testing (AUC:0.827; 95% CI: 0.802-0.857) and external validation (AUC:0.754; 95% CI: 0.739-0.771) sets. It also showed powerful ability in the prediction of bone metastasis (AUC: 0.880, 95% CI: 0.856-0.903 in the training set; AUC: 0.823, 95% CI:0.790-0.848 in the test set; AUC: 0.747, 95% CI: 0.727-0.764 in the external validation set) and lung metastasis (AUC: 0.906, 95% CI: 0.877-0.928 in training set; AUC: 0.859, 95% CI: 0.816-0.891 in the test set; AUC: 0.756, 95% CI: 0.732-0.777 in the external validation set). The AUC value of the XGB model was larger than that of nomogram in the training (0.907 vs 0.802) and external validation (0.754 vs 0.706) sets.

CONCLUSIONS

The XGB model is a better predictor of distant metastasis among MBC patients than other ML models and nomogram; furthermore, the XGB model is a powerful model for predicting bone and lung metastasis. Combining with SHAP values, it could help doctors intuitively understand the impact of each variable on outcome.

Ensemble learning based functional independence ability estimator for pediatric brain tumor survivors

A history of brain tumor strongly affects children’s cognitive abilities, performance of daily activities, quality of life, and functional outcomes. In light of the difficulties in cognition, communication, physical skills, and behavior that these patients may encounter, occupational therapists should perform a comprehensive needs-led assessment of their global functioning after recovery. Such an assessment would ensure that the patients receive adequate support and services at school, at home, and in the community. By predicting the functional activity performance of children with a history of brain tumor, clinical workers can determine the progress of their ability recovery and the optimal treatment plan. We selected several features for testing and employed common machine learning models to predict Functional Independence Measure (WeeFIM) scores. The ensemble learning models exhibited stronger predictive performance than did the individual machine learning models. The ensemble learning models effectively predicted WeeFIM scores. Machine learning models can help clinical workers predict the functional assessment scores of patients with childhood brain tumors. This study used machine learning models to predict the WeeFIM scores of patients with childhood brain tumors and to demonstrate that ensemble machine learning models are more suitable for this task than are individual machine learning models.

Role of Paper-Based Sensors in Fight against Cancer for the Developing World

Cancer is one of the major killers across the globe. According to the WHO, more than 10 million people succumbed to cancer in the year 2020 alone. The early detection of cancer is key to reducing the mortality rate. In low- and medium-income countries, the screening facilities are limited due to a scarcity of resources and equipment. Paper-based microfluidics provide a platform for a low-cost, biodegradable micro-total analysis system (µTAS) that can be used for the detection of critical biomarkers for cancer screening. This work aims to review and provide a perspective on various available paper-based methods for cancer screening. The work includes an overview of paper-based sensors, the analytes that can be detected and the detection, and readout methods used.

How Is the Lung Cancer Incidence Rate Associated with Environmental Risks? Machine-Learning-Based Modeling and Benchmarking

The lung cancer threat has become a critical issue for public health. Research has been devoted to its clinical study but only a few studies have addressed the issue from a holistic perspective that included social, economic, and environmental dimensions. Therefore, in this study, risk factors or features, such as air pollution, tobacco use, socioeconomic status, employment status, marital status, and environment, were comprehensively considered when constructing a predictive model. These risk factors were analyzed and selected using stepwise regression and the variance inflation factor to eliminate the possibility of multicollinearity. To build efficient and informative prediction models of lung cancer incidence rates, several machine learning algorithms with cross-validation were adopted, namely, linear regression, support vector regression, random forest, K-nearest neighbor, and cubist model tree. A case study in Taiwan showed that the cubist model tree with feature selection was the best model with an RMSE of 3.310 and an R-squared of 0.960. Through these predictive models, we also found that apart from smoking, the average NO2 concentration, employment percentage, and number of factories were also important factors that had significant impacts on the incidence of lung cancer. In addition, the random forest model without feature selection and with feature selection could support the interpretation of the most contributing variables. The predictive model proposed in the present study can help to precisely analyze and estimate lung cancer incidence rates so that effective preventative measures can be developed. Furthermore, the risk factors involved in the predictive model can help with the future analysis of lung cancer incidence rates from a holistic perspective.

Comparative Evaluation and Comprehensive Analysis of Machine Learning Models for Regression Problems

Artificial intelligence and machine learning applications are of significant importance almost in every field of human life to solve problems or support human experts. However, the determination of the machine learning model to achieve a superior result for a particular problem within the wide real-life application areas is still a challenging task for researchers. The success of a model could be affected by several factors such as dataset characteristics, training strategy and model responses. Therefore, a comprehensive analysis is required to determine model ability and the efficiency of the considered strategies. This study implemented ten benchmark machine learning models on seventeen varied datasets. Experiments are performed using four different training strategies 60:40, 70:30, and 80:20 hold-out and five-fold cross-validation techniques. We used three evaluation metrics to evaluate the experimental results: mean squared error, mean absolute error, and coefficient of determination (R2 score). The considered models are analyzed, and each model's advantages, disadvantages, and data dependencies are indicated. As a result of performed excess number of experiments, the deep Long-Short Term Memory (LSTM) neural network outperformed other considered models, namely, decision tree, linear regression, support vector regression with a linear and radial basis function kernels, random forest, gradient boosting, extreme gradient boosting, shallow neural network, and deep neural network. It has also been shown that cross-validation has a tremendous impact on the results of the experiments and should be considered for the model evaluation in regression studies where data mining or selection is not performed.

Breast cancer incidence and predictions (Monastir, Tunisia: 2002–2030): A registry-based study

Introduction

Breast cancer is a major public health problem worldwide. It is the leading cause of cancer deaths in females. In developing countries like Tunisia, the frequency of this cancer is still growing. The aim of this study was to determine the crude and standardized incidence rates, trends and predictions until 2030 of breast cancer incidence rates in a Tunisian governorate.

Methods

This is a descriptive study including all female patients diagnosed with breast cancer in Monastir between 2002 and 2013. The data were collected from the cancer register of the center. Tumors were coded according to the 10^th version of international classification of disease (ICD-10). Trends and predictions until 2030 were calculated using Poisson linear regression.

Results

A total of 1028 cases of female breast cancer were recorded. The median age of patients was 49 years (IQR: 41–59 years) with a minimum of 16 years and a maximum of 93 years. The age-standardized incidence rate (ASR) was of 39.12 per 100000 inhabitants. It increased significantly between 2002 and 2013 with APC of 8.4% (95% CI: 4.9; 11.9). Prediction until 2030 showed that ASR would reach 108.77 (95% CI: 57.13–209.10) per 100000 inhabitants.

Conclusion

The incidence and the chronological trends of breast cancer highlighted that this disease is of a serious concern in Tunisia. Strengthening preventive measures is a primary step to restrain its burden.

Predicting the burden of cancer in Switzerland up to 2025

Predicting the short-term evolution of the number of cancers is essential for planning investments and allocating health resources. The objective of this study was to predict the numbers of cancer cases and of the 12 most frequent cancer sites, and their age-standardized incidence rates, for the years 2019-2025 in Switzerland. Projections of the number of malignant cancer cases were obtained by combining data from two sources: forecasts of national age-standardized cancer incidence rates and population projections from the Swiss Federal Statistical Office. Age-standardized cancer incidence rates, approximating the individual cancer risk, were predicted by a low-order Autoregressive Integrated Moving Average (ARIMA) model. The contributions of changes in cancer risk (epidemiological component) and population aging and growth (demographic components) to the projected number of new cancer cases were each quantified. Between 2018 and 2025, age-standardized cancer incidence rates are predicted to stabilize for men and women at around 426 and 328/100,000, respectively (<1% change). These projected trends are expected for most cancer sites. The annual number of cancers is expected to increase from 45,676 to 52,552 (+15%), more so for men (+18%) than for women (+11%). These increases are almost entirely due to projected changes in population age structure (+12% for men and +6% for women) and population growth (+6% for both sexes). The rise in numbers of expected cancers for each site is forecast to range from 4.15% (thyroid in men) to 26% (bladder in men). While ranking of the three most frequent cancers will remain unchanged for men (1st prostate, 2nd lung, 3rd colon-rectum), colorectal cancer will overtake by 2025 lung cancer as the second most common female cancer in Switzerland, behind breast cancer. Effective and sustained prevention measures, as well as infrastructural interventions, are required to counter the increase in cancer cases and prevent any potential shortage of professionals in cancer care delivery.

Systematic Literature Review on Machine Learning and Student Performance Prediction: Critical Gaps and Possible Remedies

Improving the quality, developing and implementing systems that can provide advantages to students, and predicting students’ success during the term, at the end of the term, or in the future are some of the primary aims of education. Due to its unique ability to create relationships and obtain accurate results, artificial intelligence and machine learning are tools used in this field to achieve the expected goals. However, the diversity of studies and the differences in their content create confusion and reduce their ability to pioneer future studies. In this study, we performed a systematic literature review of student performance prediction studies in three different databases between 2010 and 2020. The results are presented as percentages by categorizing them as either model, dataset, validation, evaluation, or aims. The common points and differences in the studies are determined, and critical gaps and possible remedies are presented. The results and identified gaps could be eliminated with standardized evaluation and validation strategies. It is determined that student performance prediction studies should be more frequently focused on deep learning models in the future. Finally, the problems that can be solved using a global dataset created by a global education information consortium, as well as its advantages, are presented.

Hybrid Machine Learning Model for Body Fat Percentage Prediction Based on Support Vector Regression and Emotional Artificial Neural Networks

Obesity or excessive body fat causes multiple health problems and diseases. However, obesity treatment and control need an accurate determination of body fat percentage (BFP). The existing methods for BFP estimation require several procedures, which reduces their cost-effectivity and generalization. Therefore, developing cost-effective models for BFP estimation is vital for obesity treatment. Machine learning models, particularly hybrid models, have a strong ability to analyze challenging data and perform predictions by combining different characteristics of the models. This study proposed a hybrid machine learning model based on support vector regression and emotional artificial neural networks (SVR-EANNs) for accurate recent BFP prediction using a primary BFP dataset. SVR was applied as a consistent attribute selection model on seven properties and measurements, using the left-out sensitivity analysis, and the regression ability of the EANN was considered in the prediction phase. The proposed model was compared to seven benchmark machine learning models. The obtained results show that the proposed hybrid model (SVR-EANN) outperformed other machine learning models by achieving superior results in the three considered evaluation metrics. Furthermore, the proposed model suggested that abdominal circumference is a significant factor in BFP prediction, while age has a minor effect.