Comparison of machine learning models for predicting the risk of breast cancer-related lymphedema in Chinese women

A comprehensive review of computational diagnostic techniques for lymphedema

Lymphedema is localized swelling due to lymphatic system dysfunction, often affecting arms and legs due to fluid accumulation. It occurs in 20% to 94% of patients within 2-5 years after breast cancer treatment, with around 20% of women developing breast cancer-related lymphedema. This condition involves the accumulation of protein-rich fluid in interstitial spaces, leading to symptoms like swelling, pain, and reduced mobility that significantly impact quality of life. The early diagnosis of lymphedema helps mitigate the risk of deterioration and prevent its progression to more severe stages. Healthcare providers can reduce risks through exercise prescriptions and self-manual lymphatic drainage techniques. Lymphedema diagnosis currently relies on physical examinations and limb volume measurements, but challenges arise from a lack of standardized criteria and difficulties in detecting early stages. Recent advancements in computational imaging and decision support systems have improved diagnostic accuracy through enhanced image reconstruction and real-time data analysis. The aim of this comprehensive review is to provide an in-depth overview of the research landscape in computational diagnostic techniques for lymphedema. The computational techniques primarily include imaging-based, electrical, and machine learning (ML) approaches, which utilize advanced algorithms and data analysis. These modalities were compared based on various parameters to choose the most suitable techniques for their applications. Lymphedema detection faces challenges like subtle symptoms and inconsistent diagnostics. The research identifies bioimpedance spectroscopy (BIS), Kinect sensor and ML integration as the promising modalities for early lymphedema detection. BIS can effectively identify lymphedema as early as four months post-surgery with sensitivity of 44.1% and specificity of 95.4% in diagnosing lymphedema whereas ML and artificial neural network achieved an impressive average cross-validation accuracy of 93.75%, with sensitivity at 95.65% and specificity at 91.03%. ML and imaging can be integrated into clinical practice to enhance diagnostic accuracy and accessibility.

Predictive modeling of breast cancer-related lymphedema using machine learning algorithms

Background

Breast cancer-related lymphedema (BCRL) is one of the common complications after breast cancer surgery. It can easily lead to limb swelling, deformation and upper limb dysfunction, which has a serious impact on the physical and mental health and quality of life of patients. Previous studies have mostly used statistical methods such as linear regression and logistic regression to analyze the influencing factors, but all of them have certain limitations. Machine learning (ML) is an important branch of artificial intelligence, which can effectively overcome the problems of multivariate interaction and collinearity. This study aimed to explore the influencing factors for the occurrence of BCRL in breast cancer patients, and construct a predictive model with ML algorithms and validate its predictive value on this basis.

Methods

Clinical data of breast cancer patients admitted to Hainan Cancer Hospital from September 2018 to May 2024 were retrospectively collected. BCRL was considered as the outcome measurement, and the data were divided into training and validation sets in a ratio of 7:3. In the training set, random forest (RF), support vector machine (SVM), and eXtreme Gradient Boosting (XGBoost) algorithms were used to construct predictive models. The discrimination accuracy of the models was evaluated with receiver operating characteristic (ROC) curve analysis, sensitivity, specificity, and F1 score. The calibration of the models was assessed using calibration curves and the Hosmer-Lemeshow (H-L) Chi-squared test.

Results

Two hundred and forty patients who met the inclusion criteria were screened, and they were randomly divided into a training set (168 patients) and a validation set (72 patients) in a 7:3 ratio. In the training set, 44 cases developed BCRL, while 124 did not. There were statistically significant differences (P<0.05) in hypertension history, number of dissected lymph nodes, postoperative complications, postoperative functional exercises, chemotherapy, radiotherapy, tumor node metastasis (TNM) stage, and level of axillary lymph node dissection between the BCRL and non-BCRL groups. Among the four models, the XGBoost model showed the best predictive performance, with an area under the curve (AUC) of 0.99 in the training set and 0.89 in the validation set. The XGBoost model demonstrated good calibration in both the training and validation sets, showing good consistency with the ideal model.

Conclusions

The ML-based XGBoost model for predicting BCRL exhibits excellent performance and assists healthcare professionals in rapidly and accurately assessing the risk of BCRL occurrence.

Explainable cancer factors discovery: Shapley additive explanation for machine learning models demonstrates the best practices in the case of pancreatic cancer

Pancreatic cancer is one of digestive tract cancers with high mortality rate. Despite the wide range of available treatments and improvements in surgery, chemotherapy, and radiation therapy, the five-year prognosis for individuals diagnosed pancreatic cancer remains poor. There is still research to be done to see if immunotherapy may be used to treat pancreatic cancer. The goals of our research were to comprehend the tumor microenvironment of pancreatic cancer, found a useful biomarker to assess the prognosis of patients, and investigated its biological relevance. In this paper, machine learning methods such as random forest were fused with weighted gene co-expression networks for screening hub immune-related genes (hub-IRGs). LASSO regression model was used to further work. Thus, we got eight hub-IRGs. Based on hub-IRGs, we created a prognosis risk prediction model for PAAD that can stratify accurately and produce a prognostic risk score (IRG_Score) for each patient. In the raw data set and the validation data set, the five-year area under the curve (AUC) for this model was 0.9 and 0.7, respectively. And shapley additive explanation (SHAP) portrayed the importance of prognostic risk prediction influencing factors from a machine learning perspective to obtain the most influential certain gene (or clinical factor). The five most important factors were TRIM67, CORT, PSPN, SCAMP5, RFXAP, all of which are genes. In summary, the eight hub-IRGs had accurate risk prediction performance and biological significance, which was validated in other cancers. The result of SHAP helped to understand the molecular mechanism of pancreatic cancer.

Risk Prediction Model for Radiation-induced Dermatitis in Patients with Cervical Carcinoma Undergoing Chemoradiotherapy

PURPOSE

Radiation-induced dermatitis (RD) is a common side-effect of therapeutic ionizing radiation that can severely affect patient quality of life. This study aimed to develop a risk prediction model for the occurrence of RD in patients with cervical carcinoma undergoing chemoradiotherapy using electronic medical records (EMRs).

METHODS

Using EMRs, the clinical data of patients who underwent simultaneous radiotherapy and chemotherapy at a tertiary cancer hospital between 2017 and 2022 were retrospectively collected, and the patients were divided into two groups: a training group and a validation group. A predictive model was constructed to predict the development of RD in patients who underwent concurrent radiotherapy and chemotherapy for cervical cancer. Finally, the model's efficacy was validated using a receiver operating characteristic curve.

RESULTS

The incidence of radiation dermatitis was 89.5% (560/626) in the entire cohort, 88.6% (388/438) in the training group, and 91.5% (172/188) in the experimental group. The nomogram was established based on the following factors: age, the days between the beginning and conclusion of radiotherapy, the serum albumin after chemoradiotherapy, the use of single or multiple drugs for concurrent chemotherapy, and the total dose of afterloading radiotherapy. Internal and external verification indicated that the model had good discriminatory ability. Overall, the model achieved an area under the receiver operating characteristic curve of .66.

CONCLUSIONS

The risk of RD in patients with cervical carcinoma undergoing chemoradiotherapy is high. A risk prediction model can be developed for RD in cervical carcinoma patients undergoing chemoradiotherapy, based on over 5 years of EMR data from a tertiary cancer hospital.

Comparison of machine learning models to predict the risk of breast cancer-related lymphedema among breast cancer survivors: a cross-sectional study in China

Objective

The aim of this study was to develop and validate a series of breast cancer-related lymphoedema risk prediction models using machine learning algorithms for early identification of high-risk individuals to reduce the incidence of postoperative breast cancer lymphoedema.

Methods

This was a retrospective study conducted from January 2012 to July 2022 in a tertiary oncology hospital. Subsequent to the collection of clinical data, variables with predictive capacity for breast cancer-related lymphoedema (BCRL) were subjected to scrutiny utilizing the Least Absolute Shrinkage and Selection Operator (LASSO) technique. The entire dataset underwent a randomized partition into training and test subsets, adhering to a 7:3 distribution. Nine classification models were developed, and the model performance was evaluated based on accuracy, sensitivity, specificity, recall, precision, F-score, and area under curve (AUC) of the ROC curve. Ultimately, the selection of the optimal model hinged upon the AUC value. Grid search and 10-fold cross-validation was used to determine the best parameter setting for each algorithm.

Results

A total of 670 patients were investigated, of which 469 were in the modeling group and 201 in the validation group. A total of 174 had BCRL (25.97%). The LASSO regression model screened for the 13 features most valuable in predicting BCRL. The range of each metric in the test set for the nine models was, in order: accuracy (0.75-0.84), sensitivity (0.50-0.79), specificity (0.79-0.93), recall (0.50-0.79), precision (0.51-0.70), F score (0.56-0.69), and AUC value (0.71-0.87). Overall, LR achieved the best performance in terms of accuracy (0.81), precision (0.60), sensitivity (0.79), specificity (0.82), recall (0.79), F-score (0.68), and AUC value (0.87) for predicting BCRL.

Conclusion

The study established that the constructed logistic regression (LR) model exhibits a more favorable amalgamation of accuracy, sensitivity, specificity, recall, and AUC value. This configuration adeptly discerns patients who are at an elevated risk of BCRL. Consequently, this precise identification equips nurses with the means to undertake timely and tailored interventions, thus averting the onset of BCRL.

The application and use of artificial intelligence in cancer nursing: A systematic review

PURPOSE

Artificial Intelligence is being applied in oncology to improve patient and service outcomes. Yet, there is a limited understanding of how these advanced computational techniques are employed in cancer nursing to inform clinical practice. This review aimed to identify and synthesise evidence on artificial intelligence in cancer nursing.

METHODS

CINAHL, MEDLINE, PsycINFO, and PubMed were searched using key terms between January 2010 and December 2022. Titles, abstracts, and then full texts were screened against eligibility criteria, resulting in twenty studies being included. Critical appraisal was undertaken, and relevant data extracted and analysed. The Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines were followed.

RESULTS

Artificial intelligence was used in numerous areas including breast, colorectal, liver, and ovarian cancer care among others. Algorithms were trained and tested on primary and secondary datasets to build predictive models of health problems related to cancer. Studies reported this led to improvements in the accuracy of predicting health outcomes or identifying variables that improved outcome prediction. While nurses led most studies, few deployed an artificial intelligence based digital tool with cancer nurses in a real-world setting as studies largely focused on developing and validating predictive models.

CONCLUSION

Electronic cancer nursing datasets should be established to enable artificial intelligence techniques to be tested and if effective implemented in digital prediction and other AI-based tools. Cancer nurses need more education on machine learning and natural language processing, so they can lead and contribute to artificial intelligence developments in oncology.

Risk prediction models for breast cancer-related lymphedema: A systematic review and meta-analysis

PURPOSE

To review and critically evaluate currently available risk prediction models for breast cancer-related lymphedema (BCRL).

METHODS

PubMed, Embase, CINAHL, Scopus, Web of Science, the Cochrane Library, CNKI, SinoMed, WangFang Data, VIP Database were searched from inception to April 1, 2022, and updated on November 8, 2022. Study selection, data extraction and quality assessment were conducted by two independent reviewers. The Prediction Model Risk of Bias Assessment Tool was used to assess the risk of bias and applicability. Meta-analysis of AUC values of model external validations was performed using Stata 17.0.

RESULTS

Twenty-one studies were included, reporting twenty-two prediction models, with the AUC or C-index ranging from 0.601 to 0.965. Only two models were externally validated, with the pooled AUC of 0.70 (n = 3, 95%CI: 0.67 to 0.74), and 0.80 (n = 3, 95%CI: 0.75 to 0.86), respectively. Most models were developed using classical regression methods, with two studies using machine learning. Predictors most frequently used in included models were radiotherapy, body mass index before surgery, number of lymph nodes dissected, and chemotherapy. All studies were judged as high overall risk of bias and poorly reported.

CONCLUSIONS

Current models for predicting BCRL showed moderate to good predictive performance. However, all models were at high risk of bias and poorly reported, and their performance is probably optimistic. None of these models is suitable for recommendation in clinical practice. Future research should focus on validating, optimizing, or developing new models in well-designed and reported studies, following the methodology guidance and reporting guidelines.

Development and validation of a risk prediction model for breast cancer-related lymphedema in postoperative patients with breast cancer

OBJECTIVE

Breast cancer-related lymphedema (BCRL) is a common post-operative complication in patients with breast cancer. Here, we sought to develop and validate a predictive model of BCRL in Chinese patients with breast cancer.

METHODS

Clinical and demographic data on patients with breast cancer were collected between 2016 and 2021 at a Cancer Hospital in China. A nomogram for predicting the risk of lymphedema in postoperative patients with breast cancer was constructed and verified using R 3.5.2 software. Model performance was evaluated using area under the ROC curve (AUC) and goodness-of-fit statistics, and the model was internally validated.

RESULTS

A total of 1732 postoperative patients with breast cancer, comprising 1212 and 520 patients in the development and validation groups, respectively, were included. Of these 438 (25.39%) developed lymphedema. Significant predictors identified in the predictive model were time since breast cancer surgery, level of lymph node dissection, number of lymph nodes dissected, radiotherapy, and postoperative body mass index. At the 31.9% optimal cut-off the model had AUC values of 0.728 and 0.710 in the development and validation groups, respectively. Calibration plots showed a good match between predicted and observed rates. In decision curve analysis, the net benefit of the model was better between threshold probabilities of 10%-80%.

CONCLUSION

The model has good discrimination and accuracy for lymphedema risk assessment, which can provide a reference for individualized clinical prediction of the risk of BCRL. Multicenter prospective trials are required to verify the predictive value of the model.