Development of Machine Learning Algorithms for the Prediction of Financial Toxicity in Localized Breast Cancer Following Surgical Treatment

Interventions to mitigate cancer-related medical financial hardship: A systematic review and meta-analysis

BACKGROUND

This study systematically reviewed interventions mitigating financial hardship in patients with cancer and assessed effectiveness using a meta-analytic method.

METHODS

PubMed, Cochrane, Scopus, CINAHL, and Web of Science were searched for articles published in English during January 2000-April 2023. Two independent reviewers selected prospective clinical trials with an intervention targeting and an outcome measuring financial hardship. Quality appraisal and data extraction were performed independently by two reviewers using a quality assessment tool. A random-effects model meta-analysis was performed. Reporting followed the preferred reporting items for systematic review and meta-analyses guidelines.

RESULTS

Eleven studies (2211 participants; 55% male; mean age, 59.29 years) testing interventions including financial navigation, financial education, and cost discussion were included. Financial worry improved in only 27.3% of 11 studies. Material hardship and cost-related care nonadherence remained unchanged in the two studies measuring these outcomes. Four studies (373 participants; 37% male, mean age, 55.88 years) assessed the impact of financial navigation on financial worry using the comprehensive score of financial toxicity (COST) measure (score range, 0-44; higher score = lower financial worry) and were used for meta-analysis. There was no significant change in the mean of pooled COST score between post- and pre-intervention (1.21; 95% confidence interval, -6.54 to 8.96; p = .65). Adjusting for pre-intervention COST, mean change of COST significantly decreased by 0.88 with every 1-unit increase in pre-intervention COST (p = .02). The intervention significantly changed COST score when pre-intervention COST was ≤14.5.

CONCLUSION

A variety of interventions have been tested to mitigate financial hardship. Financial navigation can mitigate financial worry among high-risk patients.

Cancer cost-related subjective financial distress among breast cancer: a scoping review

PURPOSE

This article provided a comprehensive scoping review, synthesizing existing literature on the financial distress faced by breast cancer patients. It examined the factors contributing to financial distress, the impact on patients, coping mechanisms employed, and potential alleviation methods. The goal was to organize existing evidence and highlight possible directions for future research.

METHODS

We followed the scoping review framework proposed by the Joanna Briggs Institute (JBI) to synthesize and report evidence. We searched electronic databases, including PubMed, Web of Science, Embase, and Cochrane Library, for relevant literature. We included English articles that met the following criteria: (a) the research topic was financial distress or financial toxicity, (b) the research subjects were adult breast cancer patients, and (c) the article type was quantitative, qualitative, or mixed-methods research. We then extracted and integrated relevant information for reporting.

RESULTS

After removing duplicates, 5459 articles were retrieved, and 43 articles were included based on the inclusion and exclusion criteria. The articles addressed four main themes related to financial distress: factors associated with financial distress, impact on breast cancer patients, coping mechanisms, and potential methods for alleviation. The impact of financial distress on patients was observed in six dimensions: financial expenses, financial resources, social-psychological reactions, support seeking, coping care, and coping lifestyle. While some studies reported potential methods for alleviation, few discussed the feasibility of these solutions.

CONCLUSIONS

Breast cancer patients experience significant financial distress with multidimensional impacts. Comprehensive consideration of possible confounding factors is essential when measuring financial distress. Future research should focus on exploring and validating methods to alleviate or resolve this issue.

The usefulness of artificial intelligence in breast reconstruction: a systematic review

BACKGROUND

Artificial Intelligence (AI) offers an approach to predictive modeling. The model learns to determine specific patterns of undesirable outcomes in a dataset. Therefore, a decision-making algorithm can be built based on these patterns to prevent negative results. This systematic review aimed to evaluate the usefulness of AI in breast reconstruction.

METHODS

A systematic review was conducted in August 2022 following the Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines. MEDLINE, EMBASE, SCOPUS, and Google Scholar online databases were queried to capture all publications studying the use of artificial intelligence in breast reconstruction.

RESULTS

A total of 23 studies were full text-screened after removing duplicates, and twelve articles fulfilled our inclusion criteria. The Machine Learning algorithms applied for neuropathic pain, lymphedema diagnosis, microvascular abdominal flap failure, donor site complications associated to muscle sparing Transverse Rectus Abdominis flap, surgical complications, financial toxicity, and patient-reported outcomes after breast surgery demonstrated that AI is a helpful tool to accurately predict patient results. In addition, one study used Computer Vision technology to assist in Deep Inferior Epigastric Perforator Artery detection for flap design, considerably reducing the preoperative time compared to manual identification.

CONCLUSIONS

In breast reconstruction, AI can help the surgeon by optimizing the perioperative patients' counseling to predict negative outcomes, allowing execution of timely interventions and reducing the postoperative burden, which leads to obtaining the most successful results and improving patient satisfaction.

Use of artificial intelligence in breast surgery: a narrative review

Background and Objective

We have witnessed tremendous advances in artificial intelligence (AI) technologies. Breast surgery, a subspecialty of general surgery, has notably benefited from AI technologies. This review aims to evaluate how AI has been integrated into breast surgery practices, to assess its effectiveness in improving surgical outcomes and operational efficiency, and to identify potential areas for future research and application.

Methods

Two authors independently conducted a comprehensive search of PubMed, Google Scholar, EMBASE, and Cochrane CENTRAL databases from January 1, 1950, to September 4, 2023, employing keywords pertinent to AI in conjunction with breast surgery or cancer. The search focused on English language publications, where relevance was determined through meticulous screening of titles, abstracts, and full-texts, followed by an additional review of references within these articles. The review covered a range of studies illustrating the applications of AI in breast surgery encompassing lesion diagnosis to postoperative follow-up. Publications focusing specifically on breast reconstruction were excluded.

Key Content and Findings

AI models have preoperative, intraoperative, and postoperative applications in the field of breast surgery. Using breast imaging scans and patient data, AI models have been designed to predict the risk of breast cancer and determine the need for breast cancer surgery. In addition, using breast imaging scans and histopathological slides, models were used for detecting, classifying, segmenting, grading, and staging breast tumors. Preoperative applications included patient education and the display of expected aesthetic outcomes. Models were also designed to provide intraoperative assistance for precise tumor resection and margin status assessment. As well, AI was used to predict postoperative complications, survival, and cancer recurrence.

Conclusions

Extra research is required to move AI models from the experimental stage to actual implementation in healthcare. With the rapid evolution of AI, further applications are expected in the coming years including direct performance of breast surgery. Breast surgeons should be updated with the advances in AI applications in breast surgery to provide the best care for their patients.

Catastrophic health expenditures, insurance churn, and non-employment among women with breast cancer

BACKGROUND

Breast cancer treatment and survivorship entails a complex and expensive continuum of subspecialty care. Our objectives were to assess catastrophic health expenditures, insurance churn, and non-employment among women younger than 65 years who reported a diagnosis of breast cancer. We also evaluated changes in these outcomes related to implementation of the Affordable Care Act.

METHODS

The data source for this study was the Medical Expenditure Panel Survey (2005-2019), which is a national annual cross-sectional survey of families, providers, and insurers in the United States. To assess the impact of breast cancer, comparisons were made with a matched cohort of women without cancer. We estimated predicted marginal probabilities to quantify the effects of covariates in models for catastrophic health expenditures, insurance churn, and non-employment.

RESULTS

We identified 1490 respondents younger than 65 years who received care related to breast cancer during the study period, representing a weight-adjusted annual mean of 1 062 129 patients. Approximately 31.8% of women with breast cancer reported health expenditures in excess of 10% of their annual income. In models, the proportion of women with breast cancer who experienced catastrophic health expenditures and non-employment was inversely related to increasing income. During Affordable Care Act implementation, mean number of months of uninsurance decreased and expenditures increased among breast cancer patients.

CONCLUSIONS

Our study underscores the impact of breast cancer on financial security and opportunities for patients and their families. A multilevel understanding of these issues is needed to design effective and equitable strategies to improve quality of life and survivorship.

Financial toxicity among people with metastatic cancer: findings from the Cancer Experience Registry

PURPOSE

This study describes financial toxicity (FT) reported by people with metastatic cancer, characteristics associated with FT, and associations between FT and compensatory strategies to offset costs.

METHODS

Cancer Support Community's Cancer Experience Registry data was used to identify respondents with a solid tumor metastatic cancer who completed the Functional Assessment of Chronic Illness Therapy COmprehensive Score for Financial Toxicity (FACIT-COST) measure. Multivariable logistic regression analyses examined associations between respondent characteristics and FT, and FT and postponing medical visits, nonadherence to medications, and postponing supportive and/or psychosocial care.

RESULTS

484 individuals were included in the analysis; the most common cancers included metastatic breast (31%), lung (13%), gynecologic (10%), and colorectal (9%). Approximately half of participants (50.2%) reported some degree of FT. Those who were non-Hispanic White, Hispanic, or multiple races (compared to non-Hispanic Black), and who reported lower income, less education, and being less than one year since their cancer diagnosis had greater odds of reporting FT. Individuals with any level of FT were also more likely to report postponing medical visits (Adjusted Odds Ratio [OR] 2.58; 95% Confidence Interval [CI] 1.45-4.58), suboptimal medication adherence (Adjusted OR 5.05; 95% CI 2.77-9.20) and postponing supportive care and/or psychosocial support services (Adjusted OR 4.16; 95% CI 2.53-6.85) compared to those without FT.

CONCLUSIONS

With increases in the number of people living longer with metastatic cancer and the rising costs of therapy, there will continue to be a need to systematically screen and intervene to prevent and mitigate FT for these survivors.

Evaluation of financial toxicity and associated factors in female patients with breast cancer: a systematic review and meta-analysis

PURPOSE

These systematic review and meta-analysis were conducted to discuss the financial toxicity (FT) level among breast cancer (BC) patients and the associated demographic and economic factors.

METHODS

A systematic review and meta-analysis of single means were used by following the Joanna Briggs Institute guidelines and PRISMA guidance. Untransformed means (MRAW) were used to estimate the confidence interval for individual studies, while I2 and tau2 statistics were used to examine heterogeneity among pooled studies. Electronic databases were PubMed, CINAHL, Web of Science, Scopus, Cochrane Library, Ovid MEDLINE(R), Science Direct, and Turkish databases were used to find relevant studies published in the last 15 years (between 2008 and 2023).

RESULTS

A total of 50 studies were reviewed in the systematic review, and 11 were included in the overall and subgroup meta-analyses. The majority of reviewed studies were from the USA (38 studies), while there were four studies from China and eight studies from other countries having different types of health systems. The overall estimated FT level based on 11 pooled studies was 23.19, meaning mild level FT in the range of four categories (no FT score > 25, mild FT score 14-25, moderate FT score 1-13, and severe FT score equal to 0), with a 95% CI of 20.66-25.72. The results of subgroup meta-analyses showed that the estimated FT levels were higher among those patients who were single, with lower education levels, stage 3 patients, younger, lower income, unemployed, and living in other countries compared to those who were married, more educated, and stages 1 and 2 patients, more aged, more income, employed, and patients in the USA.

CONCLUSION

The cost-effectiveness of the treatment strategies of BC depends on the continuity of care. However, FT is one of the leading factors causing BC patients to use the required care irregularly, and it has a negative effect on adherence to treatment. So, removing the economic barriers by taking appropriate measures to decrease FT will increase the efficiency of already allocated resources to BC treatments and improve the health outcomes of BC patients.

Enhanced Surgical Decision-Making Tools in Breast Cancer: Predicting 2-Year Postoperative Physical, Sexual, and Psychosocial Well-Being following Mastectomy and Breast Reconstruction (INSPiRED 004)

BACKGROUND

We sought to predict clinically meaningful changes in physical, sexual, and psychosocial well-being for women undergoing cancer-related mastectomy and breast reconstruction 2 years after surgery using machine learning (ML) algorithms trained on clinical and patient-reported outcomes data.

PATIENTS AND METHODS

We used data from women undergoing mastectomy and reconstruction at 11 study sites in North America to develop three distinct ML models. We used data of ten sites to predict clinically meaningful improvement or worsening by comparing pre-surgical scores with 2 year follow-up data measured by validated Breast-Q domains. We employed ten-fold cross-validation to train and test the algorithms, and then externally validated them using the 11th site's data. We considered area-under-the-receiver-operating-characteristics-curve (AUC) as the primary metric to evaluate performance.

RESULTS

Overall, between 1454 and 1538 patients completed 2 year follow-up with data for physical, sexual, and psychosocial well-being. In the hold-out validation set, our ML algorithms were able to predict clinically significant changes in physical well-being (chest and upper body) (worsened: AUC range 0.69-0.70; improved: AUC range 0.81-0.82), sexual well-being (worsened: AUC range 0.76-0.77; improved: AUC range 0.74-0.76), and psychosocial well-being (worsened: AUC range 0.64-0.66; improved: AUC range 0.66-0.66). Baseline patient-reported outcome (PRO) variables showed the largest influence on model predictions.

CONCLUSIONS

Machine learning can predict long-term individual PROs of patients undergoing postmastectomy breast reconstruction with acceptable accuracy. This may better help patients and clinicians make informed decisions regarding expected long-term effect of treatment, facilitate patient-centered care, and ultimately improve postoperative health-related quality of life.

Machine learning and patient-reported outcomes for longitudinal monitoring of disease progression in metastatic breast cancer: a multicenter, retrospective analysis

BACKGROUND

Assessments of health-related quality of life (HRQoL) play an important role in transition to palliative care for women with metastatic breast cancer. We developed machine learning (ML) algorithms to analyse longitudinal HRQoL data and identify patients who may benefit from palliative care due to disease progression.

METHODS

We recruited patients from two institutions and administered the EuroQoL Visual Analog Scale (EQ-VAS) via an online platform over a 6-month period. We trained a regularised regression algorithm using 10-fold cross-validation to determine if a patient was at high or low risk of disease progression based on changes in the EQ-VAS scores using data of one institution and validated the performance on data of the other institution. Progression-free survival (PFS) was the end-point. We conducted Kaplan-Meier and Cox regression analysis adjusted for clinical risk factors.

RESULTS

Of 179 patients, 98 (54.7%) had progressive disease after a median follow-up of 14weeks. Using EQ-VAS scores collected at weeks 1-6 to predict disease progression at week 12, in the validation set (n = 63), PFS was significantly lower in the intelligent EQ-VAS high-risk versus low-risk group: median PFS 7 versus 10weeks, log-rank P < 0.038). Intelligent EQ-VAS had the strongest association with PFS (adjusted hazard ratio 2.69, 95% confidence interval 1.17-6.18, P = 0.02).

CONCLUSION

ML algorithms can analyse changes in longitudinal HRQoL data to identify patients with disease progression earlier than standard follow-up methods. Intelligent EQ-VAS scores were identified as independent prognostic factor. Future studies may validate these results to remotely monitor patients.

Financial Toxicity in Cancer Care: Implications for Clinical Care and Potential Practice Solutions

Patients with cancer face an array of financial consequences as a result of their diagnosis and treatment, collectively referred to as financial toxicity (FT). In the past 10 years, the body of literature on this subject has grown tremendously, with a recent focus on interventions and mitigation strategies. In this review, we will briefly summarize the FT literature, focusing on the contributing factors and downstream consequences on patient outcomes. In addition, we will put FT into context with our emerging understanding of the role of social determinants of health and provide a framework for understanding FT across the cancer care continuum. We will then discuss the role of the oncology community in addressing FT and outline potential strategies that oncologists and health systems can implement to reduce this undue burden on patients with cancer and their families.

The Potential of Shear Wave Elastography to Reduce Unnecessary Biopsies in Breast Cancer Diagnosis: An International, Diagnostic, Multicenter Trial

PURPOSE

 In this prospective, multicenter trial we evaluated whether additional shear wave elastography (SWE) for patients with BI-RADS 3 or 4 lesions on breast ultrasound could further refine the assessment with B-mode breast ultrasound for breast cancer diagnosis.

MATERIALS AND METHODS

 We analyzed prospective, multicenter, international data from 1288 women with breast lesions rated by conventional 2 D B-mode ultrasound as BI-RADS 3 to 4c and undergoing 2D-SWE. After reclassification with SWE the proportion of undetected malignancies should be < 2 %. All patients underwent histopathologic evaluation (reference standard).

RESULTS

 Histopathologic evaluation showed malignancy in 368 of 1288 lesions (28.6 %). The assessment with B-mode breast ultrasound resulted in 1.39 % (6 of 431) undetected malignancies (malignant lesions in BI-RADS 3) and 53.80 % (495 of 920) unnecessary biopsies (biopsies in benign lesions). Re-classifying BI-RADS 4a patients with a SWE cutoff of 2.55 m/s resulted in 1.98 % (11 of 556) undetected malignancies and a reduction of 24.24 % (375 vs. 495) of unnecessary biopsies.

CONCLUSION

 A SWE value below 2.55 m/s for BI-RADS 4a lesions could be used to downstage these lesions to follow-up, and therefore reduce the number of unnecessary biopsies by 24.24 %. However, this would come at the expense of some additionally missed cancers compared to B-mode breast ultrasound (rate of undetected malignancies 1.98 %, 11 of 556, versus 1.39 %, 6 of 431) which would, however, still be in line with the ACR BI-RADS 3 definition (< 2 % of undetected malignancies).

Machine Learning Models Using Routinely Collected Clinical Data Offer Robust and Interpretable Predictions of 90-Day Unplanned Acute Care Use for Cancer Immunotherapy Patients

PURPOSE

Clinical management of patients receiving immune checkpoint inhibitors (ICIs) could be informed using accurate predictive tools to identify patients at risk of short-term acute care utilization (ACU). We used routinely collected data to develop and assess machine learning (ML) algorithms to predict unplanned ACU within 90 days of ICI treatment initiation.

METHODS

We used aggregated electronic health record data from 7,960 patients receiving ICI treatments to train and assess eight ML algorithms. We developed the models using pre-SARS-COV-19 COVID-19 data generated between January 2016 and February 2020. We validated our algorithms using data collected between March 2020 and June 2022 (peri-COVID-19 sample). We assessed performance using area under the receiver operating characteristic curves (AUROC), sensitivity, specificity, and calibration plots. We derived intuitive explanations of predictions using variable importance and Shapley additive explanation analyses. We assessed the marginal performance of ML models compared with that of univariate and multivariate logistic regression (LR) models.

RESULTS

Most algorithms significantly outperformed the univariate and multivariate LR models. The extreme gradient boosting trees (XGBT) algorithm demonstrated the best overall performance (AUROC, 0.70; sensitivity, 0.53; specificity, 0.74) on the peri-COVID-19 sample. The algorithm performance was stable across both pre- and peri-COVID-19 samples, as well as ICI regimen and cancer groups. Type of ICI agents, oxygen saturation, diastolic blood pressure, albumin level, platelet count, immature granulocytes, absolute monocyte, chloride level, red cell distribution width, and alcohol intake were the top 10 key predictors used by the XGBT algorithm.

CONCLUSION

Machine learning algorithms trained using routinely collected data outperformed traditional statistical models when predicting 90-day ACU. The XGBT algorithm has the potential to identify high-ACU risk patients and enable preventive interventions to avoid ACU.

Machine learning models for 180-day mortality prediction of patients with advanced cancer using patient-reported symptom data

PURPOSE

The objective of the current study was to develop and test the performances of different ML algorithms which were trained using patient-reported symptom severity data to predict mortality within 180 days for patients with advanced cancer.

METHODS

We randomly selected 630 of 689 patients with advanced cancer at our institution who completed symptom PRO measures as part of routine care between 2009 and 2020. Using clinical, demographic, and PRO data, we trained and tested four ML algorithms: generalized regression with elastic net regularization (GLM), extreme gradient boosting (XGBoost) trees, support vector machines (SVM), and a single hidden layer neural network (NNET). We assessed the performance of algorithms individually as well as part of an unweighted voting ensemble on the hold-out testing sample. Performance was assessed using area under the receiver-operating characteristic curve (AUROC), sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV).

RESULTS

The starting cohort of 630 patients was randomly partitioned into training (n = 504) and testing (n = 126) samples. Of the four ML models, the XGBoost algorithm demonstrated the best performance for 180-day mortality prediction in testing data (AUROC = 0.69, sensitivity = 0.68, specificity = 0.62, PPV = 0.66, NPV = 0.64). Ensemble of all algorithms performed worst (AUROC = 0.65, sensitivity = 0.65, specificity = 0.62, PPV = 0.65, NPV = 0.62). Of individual PRO symptoms, shortness of breath emerged as the variable of highest impact on the XGBoost 180-mortality prediction (1-AUROC = 0.30).

CONCLUSION

Our findings support ML models driven by patient-reported symptom severity as accurate predictors of short-term mortality in patients with advanced cancer, highlighting the opportunity to integrate these models prospectively into future studies of goal-concordant care.

The underused potential of breast conserving therapy after neoadjuvant system treatment - Causes and solutions

Breast conserving therapy (BCT), consisting of breast conserving surgery and subsequent radiotherapy, is an equivalent option to mastectomy for women with early breast cancer. Although BCT after neoadjuvant systemic treatment (NAST) has been routinely recommend by international guidelines since many years, the rate of BCT worldwide varies largely and its potential is still underused. While the rate of BCT in western countries has increased over the past decades to currently about 70%, the rate of BCT is as low as 10% in other countries. In this review, we will evaluate the underused potential of breast conservation after NAST, identify causes, and discuss possible solutions. We identified clinical and non-clinical causes for the underuse of BCT after NAST including uncertainties within the community regarding oncologic outcomes, the correct tumor localization after NAST, the management of multifocal and multicentric tumors, margin assessment, disparities of socio-economic aspects on a patient and national level, and psychological biases affecting the shared decision-making process between patients and clinicians. Possible solutions to mitigate the underuse of BCT after NAST include interdisciplinary teams that keep the whole patient pathway in mind, optimized treatment counseling and shared decision-making, and targeted financial support to alleviate disparities.

Predicting 180-day mortality for women with ovarian cancer using machine learning and patient-reported outcome data

Contrary to national guidelines, women with ovarian cancer often receive treatment at the end of life, potentially due to the difficulty in accurately estimating prognosis. We trained machine learning algorithms to guide prognosis by predicting 180-day mortality for women with ovarian cancer using patient-reported outcomes (PRO) data. We collected data from a single academic cancer institution in the United States. Women completed biopsychosocial PRO measures every 90 days. We randomly partitioned our dataset into training and testing samples. We used synthetic minority oversampling to reduce class imbalance in the training dataset. We fitted training data to six machine learning algorithms and combined their classifications on the testing dataset into an unweighted voting ensemble. We assessed each algorithm's accuracy, sensitivity, specificity, and area under the receiver operating characteristic curve (AUROC) using testing data. We recruited 245 patients who completed 1319 PRO assessments. The final voting ensemble produced state-of-the-art results on the task of predicting 180-day mortality for ovarian cancer paitents (Accuracy = 0.79, Sensitivity = 0.71, Specificity = 0.80, AUROC = 0.76). The algorithm correctly identified 25 of the 35 women in the testing dataset who died within 180 days of assessment. Machine learning algorithms trained using PRO data offer encouraging performance in predicting whether a woman with ovarian cancer will die within 180 days. This model could be used to drive data-driven end-of-life care and address current shortcomings in care delivery. Our model demonstrates the potential of biopsychosocial PROM information to make substantial contributions to oncology prediction modeling. This model could inform clinical decision-making Future research is needed to validate these findings in a larger, more diverse sample.

Intelligent multi-modal shear wave elastography to reduce unnecessary biopsies in breast cancer diagnosis (INSPiRED 002): a retrospective, international, multicentre analysis

BACKGROUND

Breast ultrasound identifies additional carcinomas not detected in mammography but has a higher rate of false-positive findings. We evaluated whether use of intelligent multi-modal shear wave elastography (SWE) can reduce the number of unnecessary biopsies without impairing the breast cancer detection rate.

METHODS

We trained, tested, and validated machine learning algorithms using SWE, clinical, and patient information to classify breast masses. We used data from 857 women who underwent B-mode breast ultrasound, SWE, and subsequent histopathologic evaluation at 12 study sites in seven countries from 2016 to 2019. Algorithms were trained and tested on data from 11 of the 12 sites and externally validated using the additional site's data. We compared findings to the histopathologic evaluation and compared the diagnostic performance between B-mode breast ultrasound, traditional SWE, and intelligent multi-modal SWE.

RESULTS

In the external validation set (n = 285), intelligent multi-modal SWE showed a sensitivity of 100% (95% CI, 97.1-100%, 126 of 126), a specificity of 50.3% (95% CI, 42.3-58.3%, 80 of 159), and an area under the curve of 0.93 (95% CI, 0.90-0.96). Diagnostic performance was significantly higher compared to traditional SWE and B-mode breast ultrasound (P < 0.001). Unlike traditional SWE, positive-predictive values of intelligent multi-modal SWE were significantly higher compared to B-mode breast ultrasound. Unnecessary biopsies were reduced by 50.3% (79 versus 159, P < 0.001) without missing cancer compared to B-mode ultrasound.

CONCLUSION

The majority of unnecessary breast biopsies might be safely avoided by using intelligent multi-modal SWE. These results may be helpful to reduce diagnostic burden for patients, providers, and healthcare systems.

The Impact of Artificial Intelligence on Health Equity in Oncology: Scoping Review

BACKGROUND

The field of oncology is at the forefront of advances in artificial intelligence (AI) in health care, providing an opportunity to examine the early integration of these technologies in clinical research and patient care. Hope that AI will revolutionize health care delivery and improve clinical outcomes has been accompanied by concerns about the impact of these technologies on health equity.

OBJECTIVE

We aimed to conduct a scoping review of the literature to address the question, "What are the current and potential impacts of AI technologies on health equity in oncology?"

METHODS

Following PRISMA-ScR (Preferred Reporting Items for Systematic Reviews and Meta-Analyses extension for Scoping Reviews) guidelines for scoping reviews, we systematically searched MEDLINE and Embase electronic databases from January 2000 to August 2021 for records engaging with key concepts of AI, health equity, and oncology. We included all English-language articles that engaged with the 3 key concepts. Articles were analyzed qualitatively for themes pertaining to the influence of AI on health equity in oncology.

RESULTS

Of the 14,011 records, 133 (0.95%) identified from our review were included. We identified 3 general themes in the literature: the use of AI to reduce health care disparities (58/133, 43.6%), concerns surrounding AI technologies and bias (16/133, 12.1%), and the use of AI to examine biological and social determinants of health (55/133, 41.4%). A total of 3% (4/133) of articles focused on many of these themes.

CONCLUSIONS

Our scoping review revealed 3 main themes on the impact of AI on health equity in oncology, which relate to AI's ability to help address health disparities, its potential to mitigate or exacerbate bias, and its capability to help elucidate determinants of health. Gaps in the literature included a lack of discussion of ethical challenges with the application of AI technologies in low- and middle-income countries, lack of discussion of problems of bias in AI algorithms, and a lack of justification for the use of AI technologies over traditional statistical methods to address specific research questions in oncology. Our review highlights a need to address these gaps to ensure a more equitable integration of AI in cancer research and clinical practice. The limitations of our study include its exploratory nature, its focus on oncology as opposed to all health care sectors, and its analysis of solely English-language articles.

Machine learning in medicine: a practical introduction to techniques for data pre-processing, hyperparameter tuning, and model comparison

BACKGROUND

There is growing enthusiasm for the application of machine learning (ML) and artificial intelligence (AI) techniques to clinical research and practice. However, instructions on how to develop robust high-quality ML and AI in medicine are scarce. In this paper, we provide a practical example of techniques that facilitate the development of high-quality ML systems including data pre-processing, hyperparameter tuning, and model comparison using open-source software and data.

METHODS

We used open-source software and a publicly available dataset to train and validate multiple ML models to classify breast masses into benign or malignant using mammography image features and patient age. We compared algorithm predictions to the ground truth of histopathologic evaluation. We provide step-by-step instructions with accompanying code lines.

FINDINGS

Performance of the five algorithms at classifying breast masses as benign or malignant based on mammography image features and patient age was statistically equivalent (P > 0.05). Area under the receiver operating characteristics curve (AUROC) for the logistic regression with elastic net penalty was 0.89 (95% CI 0.85 - 0.94), for the Extreme Gradient Boosting Tree 0.88 (95% CI 0.83 - 0.93), for the Multivariate Adaptive Regression Spline algorithm 0.88 (95% CI 0.83 - 0.93), for the Support Vector Machine 0.89 (95% CI 0.84 - 0.93), and for the neural network 0.89 (95% CI 0.84 - 0.93).

INTERPRETATION

Our paper allows clinicians and medical researchers who are interested in using ML algorithms to understand and recreate the elements of a comprehensive ML analysis. Following our instructions may help to improve model generalizability and reproducibility in medical ML studies.

Present and future of machine learning in breast surgery: systematic review

BACKGROUND

Machine learning is a set of models and methods that can automatically detect patterns in vast amounts of data, extract information, and use it to perform decision-making under uncertain conditions. The potential of machine learning is significant, and breast surgeons must strive to be informed with up-to-date knowledge and its applications.

METHODS

A systematic database search of Embase, MEDLINE, the Cochrane database, and Google Scholar, from inception to December 2021, was conducted of original articles that explored the use of machine learning and/or artificial intelligence in breast surgery in EMBASE, MEDLINE, Cochrane database and Google Scholar.

RESULTS

The search yielded 477 articles, of which 14 studies were included in this review, featuring 73 847 patients. Four main areas of machine learning application were identified: predictive modelling of surgical outcomes; breast imaging-based context; screening and triaging of patients with breast cancer; and as network utility for detection. There is evident value of machine learning in preoperative planning and in providing information for surgery both in a cancer and an aesthetic context. Machine learning outperformed traditional statistical modelling in all studies for predicting mortality, morbidity, and quality of life outcomes. Machine learning patterns and associations could support planning, anatomical visualization, and surgical navigation.

CONCLUSION

Machine learning demonstrated promising applications for improving breast surgery outcomes and patient-centred care. Neveretheless, there remain important limitations and ethical concerns relating to implementing artificial intelligence into everyday surgical practices.

Intelligent Vacuum-Assisted Biopsy to Identify Breast Cancer Patients With Pathologic Complete Response (ypT0 and ypN0) After Neoadjuvant Systemic Treatment for Omission of Breast and Axillary Surgery

PURPOSE

Neoadjuvant systemic treatment (NST) elicits a pathologic complete response in 40%-70% of women with breast cancer. These patients may not need surgery as all local tumor has already been eradicated by NST. However, nonsurgical approaches, including imaging or vacuum-assisted biopsy (VAB), were not able to accurately identify patients without residual cancer in the breast or axilla. We evaluated the feasibility of a machine learning algorithm (intelligent VAB) to identify exceptional responders to NST.

METHODS

We trained, tested, and validated a machine learning algorithm using patient, imaging, tumor, and VAB variables to detect residual cancer after NST (ypT+ or in situ or ypN+) before surgery. We used data from 318 women with cT1-3, cN0 or +, human epidermal growth factor receptor 2-positive, triple-negative, or high-proliferative Luminal B-like breast cancer who underwent VAB before surgery (ClinicalTrials.gov identifier: NCT02948764, RESPONDER trial). We used 10-fold cross-validation to train and test the algorithm, which was then externally validated using data of an independent trial (ClinicalTrials.gov identifier: NCT02575612). We compared findings with the histopathologic evaluation of the surgical specimen. We considered false-negative rate (FNR) and specificity to be the main outcomes.

RESULTS

In the development set (n = 318) and external validation set (n = 45), the intelligent VAB showed an FNR of 0.0%-5.2%, a specificity of 37.5%-40.0%, and an area under the receiver operating characteristic curve of 0.91-0.92 to detect residual cancer (ypT+ or in situ or ypN+) after NST. Spiegelhalter's Z confirmed a well-calibrated model (z score -0.746, P = .228). FNR of the intelligent VAB was lower compared with imaging after NST, VAB alone, or combinations of both.

CONCLUSION

An intelligent VAB algorithm can reliably exclude residual cancer after NST. The omission of breast and axillary surgery for these exceptional responders may be evaluated in future trials.

An Ounce of Prediction is Worth a Pound of Cure: Risk Calculators in Breast Reconstruction

Preoperative risk calculators provide individualized risk assessment and stratification for surgical patients. Recently, several general surgery–derived models have been applied to the plastic surgery patient population, and several plastic surgery–specific calculators have been developed. In this scoping review, the authors aimed to identify and critically appraise risk calculators implemented in postmastectomy breast reconstruction.

Area Deprivation Index and Rurality in Relation to Financial Toxicity among Breast Cancer Surgical Patients: Retrospective Cross-Sectional Study of Geospatial Differences in Risk Profiles

BACKGROUND

Financial toxicity (FT) depicts the burden of cancer treatment costs and is associated with lower quality of life and survival in breast cancer patients. We examined the relationship between geospatial location, represented by rurality and Area Deprivation Index (ADI), and risk of FT.

STUDY DESIGN

A single-institution, cross-sectional study was performed on adult female surgical breast cancer patients using survey data retrospectively collected between January 2018 and June 2019. Chart reviews were used to obtain patient information, and FT was identified using the COmprehensive Score for Financial Toxicity questionnaire, which is a validated instrument. Patients' home addresses were used to determine rurality using the Rural Urban Continuum Codes and linked to national ADI score. ADI was analyzed in tertiles for univariate statistical analyses, and as a continuous variable to develop multivariable logistic regression models to evaluate the independent associations of geospatial location with FT.

RESULTS

A total of 568 surgical breast cancer patients were included. Univariate analyses found significant differences across ADI tertiles with respect to race/ethnicity, marital status, insurance type, education, and rurality. In multivariable analysis, advanced cancer stage (odds ratio [OR] 2.26, 95% CI 1.15 to 4.44) and higher ADI (OR 1.012, 95% CI 1.01 to 1.02) were associated with worsening odds of FT. Increasing age (continuous) (OR 0.976, 95% CI 0.96 to 0.99), married status (vs unmarried) (OR 0.46, 95% CI 0.30 to 0.70), and receipt of bilateral mastectomy (OR 0.56, 95% CI 0.32 to 0.96) were protective of FT.

CONCLUSIONS

FT was significantly associated with areas of greater socioeconomic deprivation as measured by the ADI. However, in adjusted analyses, rurality was not significantly associated with FT. ADI can be useful for preoperative screening of at-risk populations and the targeted deployment of community-based interventions to alleviate FT.

Breast and axillary surgery after neoadjuvant systemic treatment - A review of clinical routine recommendations and the latest clinical research

Breast and axillary surgery after neoadjuvant systemic treatment for women with breast cancer has undergone multiple paradigm changes within the past years. In this review, we provide a state-of-the-art overview of breast and axillary surgery after neoadjuvant systemic treatment from both, a clinical routine perspective and a clinical research perspective. For axillary disease, axillary lymph node dissection, sentinel lymph node biopsy, or targeted axillary dissection are nowadays recommended depending on the lymph node status before and after neoadjuvant systemic treatment. For the primary tumor in the breast, breast conserving surgery remains the standard of care. The clinical management of exceptional responders to neoadjuvant systemic treatment is a pressing knowledge gap due to the increasing number of patients who achieve a pathologic complete response to neoadjuvant systemic treatment and for whom surgery may have no therapeutic benefit. Current clinical research evaluates whether less invasive procedures can exclude residual cancer after neoadjuvant systemic treatment as reliably as surgery to possibly omit surgery for those patients in the future.

•

Breast and axillary surgery after neoadjuvant systemic treatment has evolved.

•

Choice of axillary surgery depends on lymph node status before and after treatment.

•

Optimal management of exceptional responders to neoadjuvant treatment is unclear.

•

Clinical research aims to reliably exclude residual cancer without surgery.

•

For exceptional responders, breast cancer surgery may be omitted in the future.

A New Practical Decision Rule to Better Differentiate BI-RADS 3 or 4 Breast Masses on Breast Ultrasound

OBJECTIVES

The BI-RADS classification provides a standardized way to describe ultrasound findings in breast cancer diagnostics. However, there is little information regarding which BI-RADS descriptors are most strongly associated with malignancy, to better distinguish BI-RADS 3 (follow-up imaging) and 4 (diagnostic biopsy) breast masses.

METHODS

Patients were recruited as part of an international, multicenter trial (NCT02638935). The trial enrolled 1294 women (6 excluded) categorized as BI-RADS 3 or 4 upon routine B-mode ultrasound examination. Ultrasound images were evaluated by three expert physicians according to BI-RADS. All patients underwent histopathological confirmation (reference standard). We performed univariate and multivariate analyses (chi-square test, logistic regression, and Krippendorff's alpha).

RESULTS

Histopathologic evaluation showed malignancy in 368 of 1288 masses (28.6%). Upon performing multivariate analysis, the following descriptors were significantly associated with malignancy (P < .05): age ≥50 years (OR 8.99), non-circumscribed indistinct (OR 4.05) and microlobulated margin (OR 2.95), nonparallel orientation (OR 2.69), and calcification (OR 2.64). A clinical decision rule informed by these results demonstrated a 97% sensitivity and missed fewer cancers compared to three physician experts (range of sensitivity 79-95%) and a previous decision rule (sensitivity 59%). Specificity was 44% versus 22-83%, respectively. The inter-reader reliability of the BI-RADS descriptors and of the final BI-RADS score was fair-moderate.

CONCLUSIONS

A patient should undergo a diagnostic biopsy (BI-RADS 4) instead of follow-up imaging (BI-RADS 3) if the patient is 50 years or older or exhibits at least one of the following features: calcification, nonparallel orientation of mass, non-circumscribed margin, or posterior shadowing.

The importance of multi-modal imaging and clinical information for humans and AI-based algorithms to classify breast masses (INSPiRED 003): an international, multicenter analysis

OBJECTIVES

AI-based algorithms for medical image analysis showed comparable performance to human image readers. However, in practice, diagnoses are made using multiple imaging modalities alongside other data sources. We determined the importance of this multi-modal information and compared the diagnostic performance of routine breast cancer diagnosis to breast ultrasound interpretations by humans or AI-based algorithms.

METHODS

Patients were recruited as part of a multicenter trial (NCT02638935). The trial enrolled 1288 women undergoing routine breast cancer diagnosis (multi-modal imaging, demographic, and clinical information). Three physicians specialized in ultrasound diagnosis performed a second read of all ultrasound images. We used data from 11 of 12 study sites to develop two machine learning (ML) algorithms using unimodal information (ultrasound features generated by the ultrasound experts) to classify breast masses which were validated on the remaining study site. The same ML algorithms were subsequently developed and validated on multi-modal information (clinical and demographic information plus ultrasound features). We assessed performance using area under the curve (AUC).

RESULTS

Of 1288 breast masses, 368 (28.6%) were histopathologically malignant. In the external validation set (n = 373), the performance of the two unimodal ultrasound ML algorithms (AUC 0.83 and 0.82) was commensurate with performance of the human ultrasound experts (AUC 0.82 to 0.84; p for all comparisons > 0.05). The multi-modal ultrasound ML algorithms performed significantly better (AUC 0.90 and 0.89) but were statistically inferior to routine breast cancer diagnosis (AUC 0.95, p for all comparisons ≤ 0.05).

CONCLUSIONS

The performance of humans and AI-based algorithms improves with multi-modal information.

KEY POINTS

• The performance of humans and AI-based algorithms improves with multi-modal information. • Multimodal AI-based algorithms do not necessarily outperform expert humans. • Unimodal AI-based algorithms do not represent optimal performance to classify breast masses.

The potential of combined shear wave and strain elastography to reduce unnecessary biopsies in breast cancer diagnostics - An international, multicentre trial

BACKGROUND

Shear wave elastography (SWE) and strain elastography (SE) have shown promising potential in breast cancer diagnostics by evaluating the stiffness of a lesion. Combining these two techniques could further improve the diagnostic performance. We aimed to exploratorily define the cut-offs at which adding combined SWE and SE to B-mode breast ultrasound could help reclassify Breast Imaging Reporting and Data System (BI-RADS) 3-4 lesions to reduce the number of unnecessary breast biopsies.

METHODS

We report the secondary results of a prospective, multicentre, international trial (NCT02638935). The trial enrolled 1288 women with BI-RADS 3 to 4c breast masses on conventional B-mode breast ultrasound. All patients underwent SWE and SE (index test) and histopathologic evaluation (reference standard). Reduction of unnecessary biopsies (biopsies in benign lesions) and missed malignancies after recategorising with SWE and SE were the outcome measures.

RESULTS

On performing histopathologic evaluation, 368 of 1288 breast masses were malignant. Following the routine B-mode breast ultrasound assessment, 53.80% (495 of 920 patients) underwent an unnecessary biopsy. After recategorising BI-RADS 4a lesions (SWE cut-off ≥3.70 m/s, SE cut-off ≥1.0), 34.78% (320 of 920 patients) underwent an unnecessary biopsy corresponding to a 35.35% (320 versus 495) reduction of unnecessary biopsies. Malignancies in the new BI-RADS 3 cohort were missed in 1.96% (12 of 612 patients).

CONCLUSION

Adding combined SWE and SE to routine B-mode breast ultrasound to recategorise BI-RADS 4a patients could help reduce the number of unnecessary biopsies in breast diagnostics by about 35% while keeping the rate of undetected malignancies below the 2% ACR BI-RADS 3 definition.

Vacuum-Assisted Breast Biopsy After Neoadjuvant Systemic Treatment for Reliable Exclusion of Residual Cancer in Breast Cancer Patients

Background

About 40 % of women with breast cancer achieve a pathologic complete response in the breast after neoadjuvant systemic treatment (NST). To identify these women, vacuum-assisted biopsy (VAB) was evaluated to facilitate risk-adaptive surgery. In confirmatory trials, the rates of missed residual cancer [false-negative rates (FNRs)] were unacceptably high (> 10%). This analysis aimed to improve the ability of VAB to exclude residual cancer in the breast reliably by identifying key characteristics of false-negative cases.

Methods

Uni- and multivariable logistic regressions were performed using data of a prospective multicenter trial (n = 398) to identify patient and VAB characteristics associated with false-negative cases (no residual cancer in the VAB but in the surgical specimen). Based on these findings FNR was exploratively re-calculated.

Results

In the multivariable analysis, a false-negative VAB result was significantly associated with accompanying ductal carcinoma in situ (DCIS) in the initial diagnostic biopsy [odds ratio (OR), 3.94; p < 0.001], multicentric disease on imaging before NST (OR, 2.74; p = 0.066), and age (OR, 1.03; p = 0.034). Exclusion of women with DCIS or multicentric disease (n = 114) and classication of VABs that did not remove the clip marker as uncertain representative VABs decreased the FNR to 2.9% (3/104).

Conclusion

For patients without accompanying DCIS or multicentric disease, performing a distinct representative VAB (i.e., removing a well-placed clip marker) after NST suggests that VAB might reliably exclude residual cancer in the breast without surgery. This evidence will inform the design of future trials evaluating risk-adaptive surgery for exceptional responders to NST.

Contrast of Digital and Health Literacy Between IT and Health Care Specialists Highlights the Importance of Multidisciplinary Teams for Digital Health-A Pilot Study

PURPOSE

Despite their promises, digital innovations have scarcely translated to technologies used in routine clinical practice, making the identification of barriers to successful implementation a research priority. Low levels of transdisciplinary skills represent such a barrier but so far, this has not been evaluated and compared between information technology (IT) and health care specialists. In this study, we evaluated the level of digital health literacy among IT and health care specialists.

MATERIALS AND METHODS

An anonymous questionnaire was distributed to staff at a breast cancer unit and an IT department of two German universities in December 2020. The survey questionnaire consisted of the previously validated eHealth Literacy Assessment Toolkit and additional questions with respect to age, profession, and career stage. Mann-Whitney or Wilcoxon rank-sum tests and two-sample chi-square tests were used for the analysis.

RESULTS

The survey was completed by 113 individuals: 70 (61.9%) IT specialists and 43 (38.1%) health care specialists. Health care specialists scored significantly higher on the health-related scales and IT specialists scored significantly higher on the digitally related scales. No single participant identified themselves to have the highest level of literacy on all survey questions (n = 0 of 113; 0%). Only one person (n = 1 of 113; 0.9%) consistently reported a high or the highest level of literacy.

CONCLUSION

Although IT and health care specialists showed great literacy in their respective disciplines, only few individuals combined both digital and health care literacy. Multidisciplinary teams and transdisciplinary curricula are crucial to bridge skill gaps between disciplines and to drive the implementation of digital health initiatives.