RESUMO
BACKGROUND: In the digital era when mHealth has emerged as an important venue for health care, the application of computer science, such as machine learning, has proven to be a powerful tool for health care in detecting or predicting various medical conditions by providing improved accuracy over conventional statistical or expert-based systems. Symptoms are often indicators for abnormal changes in body functioning due to illness or side effects from medical treatment. Real-time symptom report refers to the report of symptoms that patients are experiencing at the time of reporting. The use of machine learning integrating real-time patient-centered symptom report and real-time clinical analytics to develop real-time precision prediction may improve early detection of lymphedema and long term clinical decision support for breast cancer survivors who face lifelong risk of lymphedema. Lymphedema, which is associated with more than 20 distressing symptoms, is one of the most distressing and dreaded late adverse effects from breast cancer treatment. Currently there is no cure for lymphedema, but early detection can help patients to receive timely intervention to effectively manage lymphedema. Because lymphedema can occur immediately after cancer surgery or as late as 20 years after surgery, real-time detection of lymphedema using machine learning is paramount to achieve timely detection that can reduce the risk of lymphedema progression to chronic or severe stages. This study appraised the accuracy, sensitivity, and specificity to detect lymphedema status using machine learning algorithms based on real-time symptom report. METHODS: A web-based study was conducted to collect patients' real-time report of symptoms using a mHealth system. Data regarding demographic and clinical information, lymphedema status, and symptom features were collected. A total of 355 patients from 45 states in the US completed the study. Statistical and machine learning procedures were performed for data analysis. The performance of five renowned classification algorithms of machine learning were compared: Decision Tree of C4.5, Decision Tree of C5.0, gradient boosting model (GBM), artificial neural network (ANN), and support vector machine (SVM). Each classification algorithm has certain user-definable hyper parameters. Five-fold cross validation was used to optimize these hyper parameters and to choose the parameters that led to the highest average cross validation accuracy. RESULTS: Using machine leaning procedures comparing different algorithms is feasible. The ANN achieved the best performance for detecting lymphedema with accuracy of 93.75%, sensitivity of 95.65%, and specificity of 91.03%. CONCLUSIONS: A well-trained ANN classifier using real-time symptom report can provide highly accurate detection of lymphedema. Such detection accuracy is significantly higher than that achievable by current and often used clinical methods such as bio-impedance analysis. Use of a well-trained classification algorithm to detect lymphedema based on symptom features is a highly promising tool that may improve lymphedema outcomes.
RESUMO
Breast cancer-related lymphedema is a syndrome of abnormal swelling coupled with multiple symptoms resulting from obstruction or disruption of the lymphatic system associated with cancer treatment. Research has demonstrated that with increased number of symptoms reported, breast cancer survivors' limb volume increased. Lymphedema symptoms in the affected limb may indicate a latent stage of lymphedema in which changes cannot be detected by objective measures. The latent stage of lymphedema may exist months or years before overt swelling occurs. Symptom report may play an important role in detecting lymphedema in clinical practice. The purposes of this study were to: 1) examine the validity, sensitivity, and specificity of symptoms for detecting breast cancer-related lymphedema and 2) determine the best clinical cutoff point for the count of symptoms that maximized the sum of sensitivity and specificity. Data were collected from 250 women, including healthy female adults, breast cancer survivors with lymphedema, and those at risk for lymphedema. Lymphedema symptoms were assessed using a reliable and valid instrument. Validity, sensitivity, and specificity were evaluated using logistic regression, analysis of variance, and areas under receiver operating characteristic curves. Count of lymphedema symptoms was able to differentiate healthy adults from breast cancer survivors with lymphedema and those at risk for lymphedema. A diagnostic cutoff of three symptoms discriminated breast cancer survivors with lymphedema from healthy women with a sensitivity of 94% and a specificity of 97% (area under the curve =0.98). A diagnostic cutoff of nine symptoms discriminated at-risk survivors from survivors with lymphedema with a sensitivity of 64% and a specificity of 80% (area under the curve =0.72). In the absence of objective measurements capable of detecting latent stages of lymphedema, count of symptoms may be a cost-effective initial screening tool for detecting lymphedema.
RESUMO
BACKGROUND: Advances in cancer treatments continue to reduce the incidence of lymphedema. Yet, many breast cancer survivors still face long-term postoperative challenges as a result of developing lymphedema. The purpose of this study was to preliminarily evaluate The Optimal Lymph Flow program, a patient-centered education and behavioral program focusing on self-care strategies to enhance lymphedema risk reduction by promoting lymph flow and optimize body mass index (BMI). METHODS: A prospective, longitudinal, quasi-experimental design with repeated-measures was used. The study outcomes included lymph volume changes by infrared perometer, and BMI by a bioimpedance device at pre-surgery baseline, 2-4 weeks after surgery, 6-month and 12-month follow-up. A total of 140 patients were recruited and participated in The Optimal Lymph Flow program; 134 patients completed the study with 4 % attrition rate. RESULTS: Fifty-eight percent of patients had axillary node dissection and 42 % had sentinel lymph node biopsy (SLNB). The majority (97 %) of patients maintained and improved their preoperative limb volume (LV) and BMI at the study endpoint of 12 months following cancer surgery. Cumulatively, two patients with SLNB and two patients with axillary lymph node dissection had measurable lymphedema (>10 % LV change). At the 12-month follow-up, among the four patients with measurable lymphedema, two patients' LV returned to preoperative level without compression therapy but by maintaining The Optimal Lymph Flow exercises to promote daily lymph flow. CONCLUSIONS: This educational and behavioral program is effective in enhancing lymphedema risk reduction. The study provided initial evidence for emerging change in lymphedema care from treatment-focus to proactive risk reduction.