Risk-reducing surgeries for breast cancer in Brazilian patients undergoing multigene germline panel: impact of results on decision making.

PURPOSE: To assess the behavior change of high-risk breast cancer patients regarding the intention to undergo risk-reducing mastectomies (RRM) before and after genetic testing results and to identify the main influencing factors in decision-making. METHODS: Prospective cohort study conducted between November 2021 and October 2022 with women under follow-up at the high-risk outpatient clinic of the State University of Campinas (UNICAMP). Patients were referred for genetic testing, followed by counseling according to the test result. RESULTS: A total of 373 women were included. In the pre-genetic testing analysis, 54.1% of patients intended to undergo RRMs. After testing, 42.2% opted for the procedure. Behavior change occurred in 26.2%, mainly from "yes" to "no/don't know" (72,6%) (p < 0.001). The genetic test result was positive (LPV or PV) in 29.7% of patients. Among the 90 patients with positive results, 62 (68.9%) agreed to RRM, while 22 (24.4%) remained unwilling to accept RRM, regardless of the positive test. Significant influencing factors for behavior change pre- and post-genetic testing (in favor of surgery) in multivariate analysis were: positive genetic test result (OR 2.94, p < 0.001), personal cancer history (OR 2.7, p = 0.008), and ages between 40 and 49 years (OR 2.07, p = 0.008) and ≥ 50 years (OR 3.47, p < 0.001). CONCLUSION: In a Brazilian population at high-risk for breast cancer and users of the public health system, it was observed that most desired RRM, however, when genetic testing and counseling were performed, behavior change was observed, especially when the result was positive.

Identification of Potential Factors Associated with Cellulitis Following Lymphovenous Bypass Surgery in Breast Cancer Survivors.

BACKGROUND: Breast cancer is one of the most common types of cancer, with around 2.3 million cases diagnosed in 2020. One in five cancer patients develops chronic lymphedema caused by multifactorial triggers and treatment-related factors. This can lead to swelling, skin infections, and limb dysfunction, negatively affecting the patient's quality of life. This retrospective cohort study aimed to determine the associations between demographic and breast cancer characteristics and postoperative cellulitis in breast cancer survivors who underwent lymphovenous bypass surgery (LVB) at Mayo Clinic, Florida. METHODS: We performed a retrospective chart review. Data were collected retrospectively from 2016 to 2022. Sixty adult breast cancer survivors who underwent LVB were included in the final analysis based on specific inclusion and exclusion criteria. Patients were excluded if they did not meet the inclusion criteria or had incomplete follow-up data. Demographic and surgical data were extracted, including body mass index (BMI), type of anastomosis, number of anastomoses, and preoperative cellulitis status. Lymphedema measurements were performed using tape measurements. Fisher's exact test was used to determine statistically significant associations between variables and postoperative cellulitis. RESULTS: Postoperative cellulitis was more common in patients aged 60 to 69 years (43.2%), whites (75.0%), overweight or obese (90.9%), with one to four anastomoses (81.8%), and nonsmokers (79.5%). The mean International Society of Lymphology (ISL) criteria for both postoperative cellulitis and no postoperative cellulitis was 1.93. Statistically significant associations with postoperative cellulitis were found for the number of anastomoses (p = 0.021), smoking status (p = 0.049), preoperative cellulitis (p = 0.04), and the length of years with lymphedema diagnosis variable (p = 0.004). CONCLUSION: Our results suggest that a greater number of anastomoses, smoking, preoperative cellulitis, and years with lymphedema are significantly associated with an increased risk of postoperative cellulitis. Awareness of these risk factors is crucial for monitoring and early treatment of infections following surgery.

Large Language Models for Intraoperative Decision Support in Plastic Surgery: A Comparison between ChatGPT-4 and Gemini.

Background and Objectives: Large language models (LLMs) are emerging as valuable tools in plastic surgery, potentially reducing surgeons' cognitive loads and improving patients' outcomes. This study aimed to assess and compare the current state of the two most common and readily available LLMs, Open AI's ChatGPT-4 and Google's Gemini Pro (1.0 Pro), in providing intraoperative decision support in plastic and reconstructive surgery procedures. Materials and Methods: We presented each LLM with 32 independent intraoperative scenarios spanning 5 procedures. We utilized a 5-point and a 3-point Likert scale for medical accuracy and relevance, respectively. We determined the readability of the responses using the Flesch-Kincaid Grade Level (FKGL) and Flesch Reading Ease (FRE) score. Additionally, we measured the models' response time. We compared the performance using the Mann-Whitney U test and Student's t-test. Results: ChatGPT-4 significantly outperformed Gemini in providing accurate (3.59 ± 0.84 vs. 3.13 ± 0.83, p-value = 0.022) and relevant (2.28 ± 0.77 vs. 1.88 ± 0.83, p-value = 0.032) responses. Alternatively, Gemini provided more concise and readable responses, with an average FKGL (12.80 ± 1.56) significantly lower than ChatGPT-4's (15.00 ± 1.89) (p < 0.0001). However, there was no difference in the FRE scores (p = 0.174). Moreover, Gemini's average response time was significantly faster (8.15 ± 1.42 s) than ChatGPT'-4's (13.70 ± 2.87 s) (p < 0.0001). Conclusions: Although ChatGPT-4 provided more accurate and relevant responses, both models demonstrated potential as intraoperative tools. Nevertheless, their performance inconsistency across the different procedures underscores the need for further training and optimization to ensure their reliability as intraoperative decision-support tools.

Breast cancer survival after mammography dissemination in Brazil: a population-based analysis of 2,715 cases.

BACKGROUND: This study aims to assess breast cancer survival rates after one decade of mammography in a large urban area of Brazil. METHODS: It is a population-based retrospective cohort of women with breast cancer in Campinas, São Paulo, from 2010 to 2014. Age, vital status and stage were accessed through the cancer and mortality registry, and patients records. Statistics used Kaplan-Meier, log-rank and Cox's regression. RESULTS: Out of the 2,715 cases, 665 deaths (24.5%) were confirmed until early 2020. The mean age at diagnosis was 58.6 years. Women 50-69 years were 48.0%, and stage I the most frequent (25.0%). The overall mean survival was 8.4 years (8.2-8.5). The 5-year survival (5yOS) for overall, 40-49, 50-59, 60-69, 70-79 years was respectively 80.5%, 87.7%, 83.7%, 83.8% and 75.5%. The 5yOS for stages 0, I, II, III and IV was 95.2%, 92.6%, 89.4%, 71.1% and 47.1%. There was no significant difference in survival in stage I or II (p = 0.058). Compared to women 50-59 years, death's risk was 2.3 times higher for women 70-79 years and 26% lower for women 40-49 years. Concerning stage I, the risk of death was 1.5, 4.1 and 8.6 times higher, and 34% lower, respectively, for stage II, III, IV and 0. CONCLUSIONS: In Brazil, breast cancers are currently diagnosed in the early stages, although advanced cases persist. Survival rates may reflect improvements in screening, early detection and treatment. The results can reflect the current status of other regions or countries with similar health care conditions.

Breast Cancer Treatment Delay in SafetyNet Health Systems, Houston Versus Southeast Brazil.

BACKGROUND: Breast cancer outcomes among patients who use safety-net hospitals in the highly populated Harris County, Texas and Southeast Brazil are poor. It is unknown whether treatment delay contributes to these outcomes. METHODS: We conducted a retrospective cohort analysis of patients with non-metastatic breast cancer diagnosed between January 1, 2009 and December 31, 2011 at Harris Health Texas and Unicamp's Women's Hospital, Barretos Hospital, and Brazilian National Institute of Cancer, Brazil. We used Cox proportional hazards regression to evaluate association of time to treatment and risk of recurrence (ROR) or death. RESULTS: One thousand one hundred ninety-one patients were included. Women in Brazil were more frequently diagnosed with stage III disease (32.3% vs. 21.1% Texas; P = .002). Majority of patients in both populations had symptom-detected disease (63% in Brazil vs. 59% in Texas). Recurrence within 5 years from diagnosis was similar 21% versus 23%. Median time from diagnosis to first treatment defined as either systemic therapy (chemotherapy or endocrine therapy) or surgery, were comparable, 9.9 weeks versus 9.4 weeks. Treatment delay was not associated with increased ROR or death. Higher stage at diagnosis was associated with both increased ROR and death. CONCLUSION: Time from symptoms to treatment was considerably long in both populations. Treatment delay did not affect outcomes. IMPACT: Access to timely screening and diagnosis of breast cancer are priorities in these populations.

Real-world data on neoadjuvant endocrine therapy in ER-positive/HER2-negative breast cancer.

PURPOSE: Neoadjuvant endocrine therapy (NET) has been shown to be effective in ER-positive/HER2-negative breast cancer in clinical trials. However, adoption in clinical practice is still limited. Real-world data may provide useful insights into effectiveness, toxicities and quality of care, potentially rendering clinical trial results to the real-world setting. Our purpose was to report real-world data of a cohort of postmenopausal patients submitted to NET. METHODS: This prospective cohort study evaluated 146 postmenopausal female patients with ER-positive/HER2-negative breast cancer treated with NET at three tertiary hospitals between 2016 and 2018. Clinicopathological information were collected prospectively. Preoperative Endocrine Prognostic Index (PEPI) score was calculated for tumors submitted to at least 16 weeks of NET. RESULTS: Median age was 67 years old, and 87.8% had stage I-II disease. Most tumors had histological grade II (76.1%). Median pretreatment Ki67 expression was 10%. Aromatase inhibitor was used in 99.5% of patients, and median treatment duration was 21.0 weeks. No tumor progressed during NET. Breast-conserving surgery was performed in the majority of patients (63.0%), as well as sentinel lymph-node biopsy (76.7%). Pathological complete response rate was 1.0%. 43 patients (29.5%) had PEPI score 0, and 26% had PEPI scores 4-5. Posttreatment Ki67 median expression was 3.0%, and only five tumors (3.4%) showed marked increase in Ki67 expression during treatment. Seven patients (4.8%) had HER2-positive residual disease, and were treated with adjuvant chemotherapy plus trastuzumab. CONCLUSIONS: Our real-world data shows that NET is effective and safe in postmenopausal patients with ER-positive/HER2-negative breast cancer. Postmenopausal status and low-risk luminal tumor features (luminal A-like) should be used as selection criteria to ensure the best results with NET.

Prognostic assessment of breast carcinoma submitted to neoadjuvant chemotherapy with pathological non-complete response.

BACKGROUND: Breast cancer with pathological non-complete response (non-pCR) after neoadjuvant chemotherapy (NAC) has a worse prognosis. Despite Neo-Bioscore has been validated as an independent prognostic model for breast cancer submitted to NAC, non-pCR carcinoma was not assessed in this setting. METHODS: This is a retrospective trial that included women with localized breast cancer who underwent NAC and had non-pCR carcinoma in surgical specimen between 01/01/2013 to 12/31/2015 with a three-year follow-up. Survival analysis was performed by Kaplan-Meier estimator and hazard ratio (HR) set by log-rank test for the primary and secondary endpoints, respectively Disease-Free Survival (DFS) and Overall Survival (OS). According to Neo-Bioscore, the proposed prognostic model named Clustered Neo-Bioscore was classified into low (0-3), low-intermediate (4-5), high-intermediate (6) and high (7) risk. The prognostic accuracy for recurrence risk was assessed by time-dependent receiver operating characteristic (time-ROC) methodology. Multivariate Cox regression assessed the menopausal status, histological grade, Ki-67, estrogen receptor, HER2, tumor subtype, pathological and clinical stages. Confidence interval at 95% (CI95%) and statistical significance at set 2-sided p-value less than 0.05 were adopted. RESULTS: Among the 310 women enrolled, 267 patients (86.2%) had non-pCR carcinoma presenting size T3/T4 (63.3%), node-positive axilla (74.9%), stage III (62.9%), Ki-67 ≥ 20% (71.9%) and non-luminal A (78.3%). Non-pCR carcinoma presented worse DFS-3y (HR = 3.88, CI95% = 1.18-11.95) but not OS-3y (HR = 2.73, CI95% = 0.66-11.40). Clustered Neo-Bioscore discerned the recurrence risk for non-pCR carcinoma: low (DFS-3y = 0.86; baseline), low-intermediate (DFS-3y = 0.70; HR = 2.61), high-intermediate (DFS-3y = 0.13, HR = 14.05), and high (DFS-3y = not achieved; HR = 22.19). The prognostic accuracy was similar between Clustered Neo-Bioscore and Neo-Bioscore (0.76 vs 0.78, p > 0.05). Triple-negative subtype (HR = 3.6, CI95% = 1.19-10.92) and pathological stages II (HR = 5.35, CI95% = 1.19-24.01) and III (HR = 6.56, CI95% = 1.29-33.32) were prognoses for low-intermediate risk, whereas pathological stage III (HR = 13.0, CI95% = 1.60-106.10) was prognosis for low risk. CONCLUSIONS: Clustered Neo-Bioscore represents a novel prognostic model of non-pCR carcinoma undergoing NAC with a more simplified and appropriate score pattern in the assessment of prognostic factors.

Predictors of Pathological Complete Response in Women with Clinical Complete Response to Neoadjuvant Chemotherapy in Breast Carcinoma.

OBJECTIVE: There is insufficient information on predictors of pathologic complete response (pCR) to neoadjuvant chemotherapy (NAC) in breast carcinoma that also presented clinical complete response (cCR) evaluated in breast, axilla and breast and axilla. METHODS: This retrospective study included 310 women with breast carcinoma who received NAC from 1/1/13 to 12/31/15 with follow-up until 8/31/16. The factors analyzed to predict pCR and cCR were menopausal status, Ki67, estrogen receptor, histologic grade, molecular subtype, tumor size, axilla status, and stage. RESULTS: The cCR/pCR rates were 53.2/16.5% (breast), 76.3/36.8% (axilla) and 50.6/13.9% (breast and axilla). Molecular subtype and HER2-positive were independent predictors to confirm pCR in women with cCR, mainly triple negative (TN) in breast (OR 22.81, 95% CI 7.13-72.96) and breast and axilla (OR 36.06, 95% CI 8.77-148.26), but not in axilla. Ki67 ≥50% expression was predictor of cCR in breast (OR 2.00, 95% CI 1.31-3.06) and breast and axilla (OR 1.67, 95% CI 1.10-1.45). CONCLUSION: TN subtype and HER2-positive were the main independent predictors of pCR in women who also had cCR to NAC in breast and breast and axilla, but none was predictor in axilla. The Ki67 ≥50% was the independent predictor of cCR in breast and breast and axilla.

Selecting postoperative adjuvant systemic therapy for early stage breast cancer: A critical assessment of commercially available gene expression assays.

Risk stratification of patients with early stage breast cancer may support adjuvant chemotherapy decision-making. This review details the development and validation of six multi-gene classifiers, each of which claims to provide useful prognostic and possibly predictive information for early stage breast cancer patients. A careful assessment is presented of each test's analytical validity, clinical validity, and clinical utility, as well as the quality of evidence supporting its use.

Comparative Analysis of Large Language Models in Emergency Plastic Surgery Decision-Making: The Role of Physical Exam Data.

In the U.S., diagnostic errors are common across various healthcare settings due to factors like complex procedures and multiple healthcare providers, often exacerbated by inadequate initial evaluations. This study explores the role of Large Language Models (LLMs), specifically OpenAI's ChatGPT-4 and Google Gemini, in improving emergency decision-making in plastic and reconstructive surgery by evaluating their effectiveness both with and without physical examination data. Thirty medical vignettes covering emergency conditions such as fractures and nerve injuries were used to assess the diagnostic and management responses of the models. These responses were evaluated by medical professionals against established clinical guidelines, using statistical analyses including the Wilcoxon rank-sum test. Results showed that ChatGPT-4 consistently outperformed Gemini in both diagnosis and management, irrespective of the presence of physical examination data, though no significant differences were noted within each model's performance across different data scenarios. Conclusively, while ChatGPT-4 demonstrates superior accuracy and management capabilities, the addition of physical examination data, though enhancing response detail, did not significantly surpass traditional medical resources. This underscores the utility of AI in supporting clinical decision-making, particularly in scenarios with limited data, suggesting its role as a complement to, rather than a replacement for, comprehensive clinical evaluation and expertise.

AI and Ethics: A Systematic Review of the Ethical Considerations of Large Language Model Use in Surgery Research.

INTRODUCTION: As large language models receive greater attention in medical research, the investigation of ethical considerations is warranted. This review aims to explore surgery literature to identify ethical concerns surrounding these artificial intelligence models and evaluate how autonomy, beneficence, nonmaleficence, and justice are represented within these ethical discussions to provide insights in order to guide further research and practice. METHODS: A systematic review was conducted in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines. Five electronic databases were searched in October 2023. Eligible studies included surgery-related articles that focused on large language models and contained adequate ethical discussion. Study details, including specialty and ethical concerns, were collected. RESULTS: The literature search yielded 1179 articles, with 53 meeting the inclusion criteria. Plastic surgery, orthopedic surgery, and neurosurgery were the most represented surgical specialties. Autonomy was the most explicitly cited ethical principle. The most frequently discussed ethical concern was accuracy (n = 45, 84.9%), followed by bias, patient confidentiality, and responsibility. CONCLUSION: The ethical implications of using large language models in surgery are complex and evolving. The integration of these models into surgery necessitates continuous ethical discourse to ensure responsible and ethical use, balancing technological advancement with human dignity and safety.

AI in Hand Surgery: Assessing Large Language Models in the Classification and Management of Hand Injuries.

Background: OpenAI's ChatGPT (San Francisco, CA, USA) and Google's Gemini (Mountain View, CA, USA) are two large language models that show promise in improving and expediting medical decision making in hand surgery. Evaluating the applications of these models within the field of hand surgery is warranted. This study aims to evaluate ChatGPT-4 and Gemini in classifying hand injuries and recommending treatment. Methods: Gemini and ChatGPT were given 68 fictionalized clinical vignettes of hand injuries twice. The models were asked to use a specific classification system and recommend surgical or nonsurgical treatment. Classifications were scored based on correctness. Results were analyzed using descriptive statistics, a paired two-tailed t-test, and sensitivity testing. Results: Gemini, correctly classifying 70.6% hand injuries, demonstrated superior classification ability over ChatGPT (mean score 1.46 vs. 0.87, p-value < 0.001). For management, ChatGPT demonstrated higher sensitivity in recommending surgical intervention compared to Gemini (98.0% vs. 88.8%), but lower specificity (68.4% vs. 94.7%). When compared to ChatGPT, Gemini demonstrated greater response replicability. Conclusions: Large language models like ChatGPT and Gemini show promise in assisting medical decision making, particularly in hand surgery, with Gemini generally outperforming ChatGPT. These findings emphasize the importance of considering the strengths and limitations of different models when integrating them into clinical practice.

Artificial-Intelligence-Based Clinical Decision Support Systems in Primary Care: A Scoping Review of Current Clinical Implementations.

Primary Care Physicians (PCPs) are the first point of contact in healthcare. Because PCPs face the challenge of managing diverse patient populations while maintaining up-to-date medical knowledge and updated health records, this study explores the current outcomes and effectiveness of implementing Artificial Intelligence-based Clinical Decision Support Systems (AI-CDSSs) in Primary Healthcare (PHC). Following the PRISMA-ScR guidelines, we systematically searched five databases, PubMed, Scopus, CINAHL, IEEE, and Google Scholar, and manually searched related articles. Only CDSSs powered by AI targeted to physicians and tested in real clinical PHC settings were included. From a total of 421 articles, 6 met our criteria. We found AI-CDSSs from the US, Netherlands, Spain, and China whose primary tasks included diagnosis support, management and treatment recommendations, and complication prediction. Secondary objectives included lessening physician work burden and reducing healthcare costs. While promising, the outcomes were hindered by physicians' perceptions and cultural settings. This study underscores the potential of AI-CDSSs in improving clinical management, patient satisfaction, and safety while reducing physician workload. However, further work is needed to explore the broad spectrum of applications that the new AI-CDSSs have in several PHC real clinical settings and measure their clinical outcomes.

Identification of Pain through Actigraphy-Recorded Patient Movement: A Comprehensive Review.

Chronic pain affects over 50 million people in the United States, particularly older adults, making effective assessment and treatment essential in primary care. Actigraphy, which monitors and records limb movement to estimate wakefulness and sleep, has emerged as a valuable tool for assessing pain by providing insights into activity patterns. This review highlights the non-invasive, cost-effective nature of actigraphy in pain monitoring, along with its ability to offer continuous, detailed data on patient movement. However, actigraphy's reliance on physical activity as a proxy for pain, and its inability to directly measure pain intensity, limit its applicability to certain pain types, such as neuropathic pain. Further research is needed to overcome these limitations and to improve the effectiveness of actigraphy in diverse clinical settings.

Artificial Intelligence in Postoperative Care: Assessing Large Language Models for Patient Recommendations in Plastic Surgery.

Since their release, the medical community has been actively exploring large language models' (LLMs) capabilities, which show promise in providing accurate medical knowledge. One potential application is as a patient resource. This study analyzes and compares the ability of the currently available LLMs, ChatGPT-3.5, GPT-4, and Gemini, to provide postoperative care recommendations to plastic surgery patients. We presented each model with 32 questions addressing common patient concerns after surgical cosmetic procedures and evaluated the medical accuracy, readability, understandability, and actionability of the models' responses. The three LLMs provided equally accurate information, with GPT-3.5 averaging the highest on the Likert scale (LS) (4.18 ± 0.93) (p = 0.849), while Gemini provided significantly more readable (p = 0.001) and understandable responses (p = 0.014; p = 0.001). There was no difference in the actionability of the models' responses (p = 0.830). Although LLMs have shown their potential as adjunctive tools in postoperative patient care, further refinement and research are imperative to enable their evolution into comprehensive standalone resources.

Clinical and Surgical Applications of Large Language Models: A Systematic Review.

Background: Large language models (LLMs) represent a recent advancement in artificial intelligence with medical applications across various healthcare domains. The objective of this review is to highlight how LLMs can be utilized by clinicians and surgeons in their everyday practice. Methods: A systematic review was conducted following the Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines. Six databases were searched to identify relevant articles. Eligibility criteria emphasized articles focused primarily on clinical and surgical applications of LLMs. Results: The literature search yielded 333 results, with 34 meeting eligibility criteria. All articles were from 2023. There were 14 original research articles, four letters, one interview, and 15 review articles. These articles covered a wide variety of medical specialties, including various surgical subspecialties. Conclusions: LLMs have the potential to enhance healthcare delivery. In clinical settings, LLMs can assist in diagnosis, treatment guidance, patient triage, physician knowledge augmentation, and administrative tasks. In surgical settings, LLMs can assist surgeons with documentation, surgical planning, and intraoperative guidance. However, addressing their limitations and concerns, particularly those related to accuracy and biases, is crucial. LLMs should be viewed as tools to complement, not replace, the expertise of healthcare professionals.

Comparative Analysis of Artificial Intelligence Virtual Assistant and Large Language Models in Post-Operative Care.

In postoperative care, patient education and follow-up are pivotal for enhancing the quality of care and satisfaction. Artificial intelligence virtual assistants (AIVA) and large language models (LLMs) like Google BARD and ChatGPT-4 offer avenues for addressing patient queries using natural language processing (NLP) techniques. However, the accuracy and appropriateness of the information vary across these platforms, necessitating a comparative study to evaluate their efficacy in this domain. We conducted a study comparing AIVA (using Google Dialogflow) with ChatGPT-4 and Google BARD, assessing the accuracy, knowledge gap, and response appropriateness. AIVA demonstrated superior performance, with significantly higher accuracy (mean: 0.9) and lower knowledge gap (mean: 0.1) compared to BARD and ChatGPT-4. Additionally, AIVA's responses received higher Likert scores for appropriateness. Our findings suggest that specialized AI tools like AIVA are more effective in delivering precise and contextually relevant information for postoperative care compared to general-purpose LLMs. While ChatGPT-4 shows promise, its performance varies, particularly in verbal interactions. This underscores the importance of tailored AI solutions in healthcare, where accuracy and clarity are paramount. Our study highlights the necessity for further research and the development of customized AI solutions to address specific medical contexts and improve patient outcomes.

Evaluating Large Language Model (LLM) Performance on Established Breast Classification Systems.

Medical researchers are increasingly utilizing advanced LLMs like ChatGPT-4 and Gemini to enhance diagnostic processes in the medical field. This research focuses on their ability to comprehend and apply complex medical classification systems for breast conditions, which can significantly aid plastic surgeons in making informed decisions for diagnosis and treatment, ultimately leading to improved patient outcomes. Fifty clinical scenarios were created to evaluate the classification accuracy of each LLM across five established breast-related classification systems. Scores from 0 to 2 were assigned to LLM responses to denote incorrect, partially correct, or completely correct classifications. Descriptive statistics were employed to compare the performances of ChatGPT-4 and Gemini. Gemini exhibited superior overall performance, achieving 98% accuracy compared to ChatGPT-4's 71%. While both models performed well in the Baker classification for capsular contracture and UTSW classification for gynecomastia, Gemini consistently outperformed ChatGPT-4 in other systems, such as the Fischer Grade Classification for gender-affirming mastectomy, Kajava Classification for ectopic breast tissue, and Regnault Classification for breast ptosis. With further development, integrating LLMs into plastic surgery practice will likely enhance diagnostic support and decision making.

Artificial Intelligence Support for Informal Patient Caregivers: A Systematic Review.

This study aims to explore how artificial intelligence can help ease the burden on caregivers, filling a gap in current research and healthcare practices due to the growing challenge of an aging population and increased reliance on informal caregivers. We conducted a search with Google Scholar, PubMed, Scopus, IEEE Xplore, and Web of Science, focusing on AI and caregiving. Our inclusion criteria were studies where AI supports informal caregivers, excluding those solely for data collection. Adhering to PRISMA 2020 guidelines, we eliminated duplicates and screened for relevance. From 947 initially identified articles, 10 met our criteria, focusing on AI's role in aiding informal caregivers. These studies, conducted between 2012 and 2023, were globally distributed, with 80% employing machine learning. Validation methods varied, with Hold-Out being the most frequent. Metrics across studies revealed accuracies ranging from 71.60% to 99.33%. Specific methods, like SCUT in conjunction with NNs and LibSVM, showcased accuracy between 93.42% and 95.36% as well as F-measures spanning 93.30% to 95.41%. AUC values indicated model performance variability, ranging from 0.50 to 0.85 in select models. Our review highlights AI's role in aiding informal caregivers, showing promising results despite different approaches. AI tools provide smart, adaptive support, improving caregivers' effectiveness and well-being.

The impact of the COVID-19 pandemic on breast cancer screening and diagnosis in a Brazilian metropolitan area.

OBJECTIVES: To evaluate the performance of breast cancer screening and early diagnosis during the pandemic, compared to the pre-pandemic period.Setting: The public referral centre for screening in Campinas, São Paulo State, Brazil. METHODS: This is an audit study of performance screening and diagnostic indicators. Two periods were analysed: 2019, the pre-COVID period, and 2020, the COVID period. All women who underwent mammography in these periods were included. Indicators were compared between periods, and the US Breast Cancer Surveillance Consortium benchmarks were used as a reference. RESULTS: A comparison between the periods shows a reduction of 57.4% in screening and 4.4% in diagnosis using mammography. Cancer detection rate per 1000 screening mammograms dropped from 4.62 to 2.83 (p = 0.031), while it increased from 84.43 to 89.36 in diagnosis mammograms (p = 0.701), higher than the reference (34.4, p < 0.001). With regard to diagnosis, the proportion of minimal cancers was reduced (p = 0.005) and was lower than the reference (40.0%, p < 0.001), along with the proportion of node-negative invasive cancers (p < 0.001). The mean size of invasive tumours was similar in the two periods (32.50 mm and 33.40 mm, p = 0.808) but larger than the reference value (16.50 mm, p < 0.001). Recall rate was lower in the COVID period (22.55% vs. 27.37%, p = 0.015). CONCLUSION: The COVID pandemic caused an overall decrease in breast screening and detection of breast cancer cases, although the reduction in number of diagnosis mammograms performed was minimal. Tumour mean size was large in both periods, the pandemic highlighting a previous profile of detection at an advanced stage.