Individualized survival prediction and surgery recommendation for patients with glioblastoma.

Background: There is a lack of individualized evidence on surgical choices for glioblastoma (GBM) patients. Aim: This study aimed to make individualized treatment recommendations for patients with GBM and to determine the importance of demographic and tumor characteristic variables in the selection of extent of resection. Methods: We proposed Balanced Decision Ensembles (BDE) to make survival predictions and individualized treatment recommendations. We developed several DL models to counterfactually predict the individual treatment effect (ITE) of patients with GBM. We divided the patients into the recommended (Rec.) and anti-recommended groups based on whether their actual treatment was consistent with the model recommendation. Results: The BDE achieved the best recommendation effects (difference in restricted mean survival time (dRMST): 5.90; 95% confidence interval (CI), 4.40-7.39; hazard ratio (HR): 0.71; 95% CI, 0.65-0.77), followed by BITES and DeepSurv. Inverse probability treatment weighting (IPTW)-adjusted HR, IPTW-adjusted OR, natural direct effect, and control direct effect demonstrated better survival outcomes of the Rec. group. Conclusion: The ITE calculation method is crucial, as it may result in better or worse recommendations. Furthermore, the significant protective effects of machine recommendations on survival time and mortality indicate the superiority of the model for application in patients with GBM. Overall, the model identifies patients with tumors located in the right and left frontal and middle temporal lobes, as well as those with larger tumor sizes, as optimal candidates for SpTR.

Determining individual suitability for neoadjuvant systemic therapy in breast cancer patients through deep learning.

BACKGROUND: The survival advantage of neoadjuvant systemic therapy (NST) for breast cancer patients remains controversial, especially when considering the heterogeneous characteristics of individual patients. OBJECTIVE: To discern the variability in responses to breast cancer treatment at the individual level and propose personalized treatment recommendations utilizing deep learning (DL). METHODS: Six models were developed to offer individualized treatment suggestions. Outcomes for patients whose actual treatments aligned with model recommendations were compared to those whose did not. The influence of certain baseline features of patients on NST selection was visualized and quantified by multivariate logistic regression and Poisson regression analyses. RESULTS: Our study included 94,487 female breast cancer patients. The Balanced Individual Treatment Effect for Survival data (BITES) model outperformed other models in performance, showing a statistically significant protective effect with inverse probability treatment weighting (IPTW)-adjusted baseline features [IPTW-adjusted hazard ratio: 0.51, 95% confidence interval (CI), 0.41-0.64; IPTW-adjusted risk difference: 21.46, 95% CI 18.90-24.01; IPTW-adjusted difference in restricted mean survival time: 21.51, 95% CI 19.37-23.80]. Adherence to BITES recommendations is associated with reduced breast cancer mortality and fewer adverse effects. BITES suggests that patients with TNM stage IIB, IIIB, triple-negative subtype, a higher number of positive axillary lymph nodes, and larger tumors are most likely to benefit from NST. CONCLUSIONS: Our results demonstrated the potential of BITES to aid in clinical treatment decisions and offer quantitative treatment insights. In our further research, these models should be validated in clinical settings and additional patient features as well as outcome measures should be studied in depth.

Optimizing adjuvant treatment options for patients with glioblastoma.

Background: This study focused on minimizing the costs and toxic effects associated with unnecessary chemotherapy. We sought to optimize the adjuvant therapy strategy, choosing between radiotherapy (RT) and chemoradiotherapy (CRT), for patients based on their specific characteristics. This selection process utilized an innovative deep learning method. Methods: We trained six machine learning (ML) models to advise on the most suitable treatment for glioblastoma (GBM) patients. To assess the protective efficacy of these ML models, we employed various metrics: hazards ratio (HR), inverse probability treatment weighting (IPTW)-adjusted HR (HRa), the difference in restricted mean survival time (dRMST), and the number needed to treat (NNT). Results: The Balanced Individual Treatment Effect for Survival data (BITES) model emerged as the most effective, demonstrating significant protective benefits (HR: 0.53, 95% CI, 0.48-0.60; IPTW-adjusted HR: 0.65, 95% CI, 0.55-0.78; dRMST: 7.92, 95% CI, 7.81-8.15; NNT: 1.67, 95% CI, 1.24-2.41). Patients whose treatment aligned with BITES recommendations exhibited notably better survival rates compared to those who received different treatments, both before and after IPTW adjustment. In the CRT-recommended group, a significant survival advantage was observed when choosing CRT over RT (p < 0.001). However, this was not the case in the RT-recommended group (p = 0.06). Males, older patients, and those whose tumor invasion is confined to the ventricular system were more frequently advised to undergo RT. Conclusion: Our study suggests that BITES can effectively identify GBM patients likely to benefit from CRT. These ML models show promise in transforming the complex heterogeneity of real-world clinical practice into precise, personalized treatment recommendations.

Quantified treatment effect at the individual level is more indicative for personalized radical prostatectomy recommendation: implications for prostate cancer treatment using deep learning.

BACKGROUND: There are potential uncertainties and overtreatment existing in radical prostatectomy (RP) for prostate cancer (PCa) patients, thus identifying optimal candidates is quite important. PURPOSE: This study aims to establish a novel causal inference deep learning (DL) model to discern whether a patient can benefit more from RP and to identify heterogeneity in treatment responses among PCa patients. METHODS: We introduce the Self-Normalizing Balanced individual treatment effect for survival data (SNB). Six models were trained to make individualized treatment recommendations for PCa patients. Inverse probability treatment weighting (IPTW) was used to avoid treatment selection bias. RESULTS: 35,236 patients were included. Patients whose actual treatment was consistent with SNB recommendations had better survival outcomes than those who were inconsistent (multivariate hazard ratio (HR): 0.76, 95% confidence interval (CI), 0.64-0.92; IPTW-adjusted HR: 0.77, 95% CI, 0.61-0.95; risk difference (RD): 3.80, 95% CI, 2.48-5.11; IPTW-adjusted RD: 2.17, 95% CI, 0.92-3.35; the difference in restricted mean survival time (dRMST): 3.81, 95% CI, 2.66-4.85; IPTW-adjusted dRMST: 3.23, 95% CI, 2.06-4.45). Keeping other covariates unchanged, patients with 1 ng/mL increase in PSA levels received RP caused 1.77 months increase in the time to 90% mortality, and the similar results could be found in age, Gleason score, tumor size, TNM stages, and metastasis status. CONCLUSIONS: Our highly interpretable and reliable DL model (SNB) may identify patients with PCa who could benefit from RP, outperforming other models and clinical guidelines. Additionally, the DL-based treatment guidelines obtained can provide priori evidence for subsequent studies.

Deep learning-guided adjuvant chemotherapy selection for elderly patients with breast cancer.

PURPOSE: The efficacy of adjuvant chemotherapy in elderly breast cancer patients is currently controversial. This study aims to provide personalized adjuvant chemotherapy recommendations using deep learning (DL). METHODS: Six models with various causal inference approaches were trained to make individualized chemotherapy recommendations. Patients who received actual treatment recommended by DL models were compared with those who did not. Inverse probability treatment weighting (IPTW) was used to reduce bias. Linear regression, IPTW-adjusted risk difference (RD), and SurvSHAP(t) were used to interpret the best model. RESULTS: A total of 5352 elderly breast cancer patients were included. The median (interquartile range) follow-up time was 52 (30-80) months. Among all models, the balanced individual treatment effect for survival data (BITES) performed best. Treatment according to following BITES recommendations was associated with survival benefit, with a multivariate hazard ratio (HR) of 0.78 (95% confidence interval (CI): 0.64-0.94), IPTW-adjusted HR of 0.74 (95% CI: 0.59-0.93), RD of 12.40% (95% CI: 8.01-16.90%), IPTW-adjusted RD of 11.50% (95% CI: 7.16-15.80%), difference in restricted mean survival time (dRMST) of 12.44 (95% CI: 8.28-16.60) months, IPTW-adjusted dRMST of 7.81 (95% CI: 2.93-11.93) months, and p value of the IPTW-adjusted Log-rank test of 0.033. By interpreting BITES, the debiased impact of patient characteristics on adjuvant chemotherapy was quantified, which mainly included breast cancer subtype, tumor size, number of positive lymph nodes, TNM stages, histological grades, and surgical type. CONCLUSION: Our results emphasize the potential of DL models in guiding adjuvant chemotherapy decisions for elderly breast cancer patients.

Reasoning and causal inference regarding surgical options for patients with low-grade gliomas using machine learning: A SEER-based study.

BACKGROUND: Due to the heterogeneity of low-grade gliomas (LGGs), the lack of randomized control trials, and strong clinical evidence, the effect of the extent of resection (EOR) is currently controversial. AIM: To determine the best choice between subtotal resection (STR) and gross-total resection (GTR) for individual patients and to identify features that are potentially relevant to treatment heterogeneity. METHODS: Patients were enrolled from the SEER database. We used a novel DL approach to make treatment recommendations for patients with LGG. We also made causal inference of the average treatment effect (ATE) of GTR compared with STR. RESULTS: The patients were divided into the Consis. and In-consis. groups based on whether their actual treatment and model recommendations were consistent. Better brain cancer-specific survival (BCSS) outcomes in the Consis. group was observed. Overall, we also identified two subgroups that showed strong heterogeneity in response to GTR. By interpreting the models, we identified numerous variables that may be related to treatment heterogeneity. CONCLUSIONS: This is the first study to infer the individual treatment effect, make treatment recommendation, and guide surgical options through deep learning approach in LGG research. Through causal inference, we found that heterogeneous responses to STR and GTR exist in patients with LGG. Visualization of the model yielded several factors that contribute to treatment heterogeneity, which are worthy of further discussion.

Effective early strategy to prevent olfactory and gustatory dysfunction in COVID-19: A randomized controlled trial.

BACKGROUND: Olfactory and gustatory dysfunctions (OGDs) are key symptoms of COVID-19, which may lead to neurological complications, and lack of effective treatment. This may be because post-disease treatments may be too late to protect the olfactory and gustatory functions. AIM: To evaluate the effectiveness of early use of saline nasal irrigation (SNI), corticosteroid nasal spray, and saline or chlorhexidine gluconate mouthwash for preventing OGDs in COVID-19. DESIGN: This study was a double-blind randomized controlled trial. METHODS: The study was conducted from May 5 to June 16, 2022. We recruited patients from three hospitals who were admitted with COVID-19 but without OGDs on the day of admission. Olfactory and gustatory functions were evaluated using the Taste and Smell Survey and the numerical visual analog scale. Participants were randomized to the saline, drug, or control groups. The control group received no intervention, saline group received SNI plus saline nasal spray and mouthwash, and the trial group received SNI plus budesonide nasal spray and chlorhexidine gluconate mouthwash. Participants were assessed again on the day of discharge. RESULTS: A total of 379 patients completed the trial. The prevalence of OGDs was significantly lower in the saline (11.8%, 95% CI, 6.6-19.0%; P < 0.001) and trial (8.3%, 95% CI, 4.1-14.8%; P < 0.001) groups than in the control group (40.0%, 95% CI, 31.8-48.6%). Additionally, both interventions reduced the severity of OGDs. CONCLUSIONS: We demonstrated effective strategies for preventing COVID-19-related OGDs, and the findings may guide early management of SARS-CoV-2 infection to reduce the incidence of COVID-19-related complications.

Analyzing and predicting the risk of death in stroke patients using machine learning.

Background: Stroke is an acute disorder and dysfunction of the focal neurological system that has long been recognized as one of the leading causes of death and severe disability in most regions globally. This study aimed to supplement and exploit multiple comorbidities, laboratory tests and demographic factors to more accurately predict death related to stroke, and furthermore, to make inferences about the heterogeneity of treatment in stroke patients to guide better treatment planning. Methods: We extracted data from the Medical Information Mart from the Intensive Care (MIMIC)-IV database. We compared the distribution of the demographic factors between the control and death groups. Subsequently, we also developed machine learning (ML) models to predict mortality among stroke patients. Furthermore, we used meta-learner to recognize the heterogeneity effects of warfarin and human albumin. We comprehensively evaluated and interpreted these models using Shapley Additive Explanation (SHAP) analysis. Results: We included 7,483 patients with MIMIC-IV in this study. Of these, 1,414 (18.9%) patients died during hospitalization or 30 days after discharge. We found that the distributions of age, marital status, insurance type, and BMI differed between the two groups. Our machine learning model achieved the highest level of accuracy to date in predicting mortality in stroke patients. We also observed that patients who were consistent with the model determination had significantly better survival outcomes than the inconsistent population and were better than the overall treatment group. Conclusion: We used several highly interpretive machine learning models to predict stroke prognosis with the highest accuracy to date and to identify heterogeneous treatment effects of warfarin and human albumin in stroke patients. Our interpretation of the model yielded a number of findings that are consistent with clinical knowledge and warrant further study and verification.

Alterations in smell or taste in individuals infected with SARS-CoV-2 during periods of Omicron variant dominance.

OBJECTIVES: To characterize the prevalence, severity, correlation with initial symptoms, and role of vaccination in patients with COVID-19 with smell or taste alterations (STAs). METHODS: We conducted an observational study of patients infected with SARS-CoV-2 Omicron admitted to three hospitals between May 17 and June 16, 2022. The olfactory and gustatory functions were evaluated using the taste and smell survey and the numerical visual analog scale at two time points. RESULTS: The T1 and T2 time point assessments were completed by 688 and 385 participants, respectively. The prevalence of STAs at two time points was 41.3% vs 42.6%. Furthermore, no difference existed in the severity distribution of taste and smell survey, smell, or taste visual analog scale scores between the groups. Patients with initial symptoms of headache (P = 0.03) and muscle pain (P = 0.04) were more likely to develop STAs, whereas higher education; three-dose vaccination; no symptoms yet; or initial symptoms of cough, throat discomfort, and fever demonstrated protective effects, and the results were statistically significant. CONCLUSION: The prevalence of STAs did not decrease significantly during the Omicron dominance, but the severity was reduced, and vaccination demonstrated a protective effect. In addition, the findings suggest that the presence of STAs is likely to be an important indicator of viral invasion of the nervous system.

Prediction of acute kidney injury in patients with femoral neck fracture utilizing machine learning.

Background: Acute kidney injury (AKI) is a common complication associated with significant morbidity and mortality in high-energy trauma patients. Given the poor efficacy of interventions after AKI development, it is important to predict AKI before its diagnosis. Therefore, this study aimed to develop models using machine learning algorithms to predict the risk of AKI in patients with femoral neck fractures. Methods: We developed machine-learning models using the Medical Information Mart from Intensive Care (MIMIC)-IV database. AKI was predicted using 10 predictive models in three-time windows, 24, 48, and 72 h. Three optimal models were selected according to the accuracy and area under the receiver operating characteristic curve (AUROC), and the hyperparameters were adjusted using a random search algorithm. The Shapley additive explanation (SHAP) analysis was used to determine the impact and importance of each feature on the prediction. Compact models were developed using important features chosen based on their SHAP values and clinical availability. Finally, we evaluated the models using metrics such as accuracy, precision, AUROC, recall, F1 scores, and kappa values on the test set after hyperparameter tuning. Results: A total of 1,596 patients in MIMIC-IV were included in the final cohort, and 402 (25%) patients developed AKI after surgery. The light gradient boosting machine (LightGBM) model showed the best overall performance for predicting AKI before 24, 48, and 72 h. AUROCs were 0.929, 0.862, and 0.904. The SHAP value was used to interpret the prediction models. Renal function markers and perioperative blood transfusions are the most critical features for predicting AKI. In compact models, LightGBM still performs the best. AUROCs were 0.930, 0.859, and 0.901. Conclusions: In our analysis, we discovered that LightGBM had the best metrics among all algorithms used. Our study identified the LightGBM as a solid first-choice algorithm for early AKI prediction in patients after femoral neck fracture surgery.

The potency of psychiatric questionnaires to distinguish major mental disorders in Chinese outpatients.

Background: Considering the huge population in China, the available mental health resources are inadequate. Thus, our study aimed to evaluate whether mental questionnaires, serving as auxiliary diagnostic tools, have efficient diagnostic ability in outpatient psychiatric services. Methods: We conducted a retrospective study of Chinese psychiatric outpatients. Altogether 1,182, 5,069, and 4,958 records of Symptom Checklist-90 (SCL-90), Hamilton Anxiety Rating Scale (HAM-A), and Hamilton Depression Rating Scale (HAM-D), respectively, were collected from March 2021 to July 2022. The Mann-Whitney U test was applied to subscale scores and total scores of SCL-90, HAM-A, and HAM-D between the two sexes (male and female groups), different age groups, and four diagnostic groups (anxiety disorder, depressive disorder, bipolar disorder, and schizophrenia). Kendall's tau coefficient analysis and machine learning were also conducted in the diagnostic groups. Results: We found significant differences in most subscale scores for both age and gender groups. Using the Mann-Whitney U test and Kendall's tau coefficient analysis, we found that there were no statistically significant differences in diseases in total scale scores and nearly all subscale scores. The results of machine learning (ML) showed that for HAM-A, anxiety had a small degree of differentiation with an AUC of 0.56, while other diseases had an AUC close to 0.50. As for HAM-D, bipolar disorder was slightly distinguishable with an AUC of 0.60, while the AUC of other diseases was lower than 0.50. In SCL-90, all diseases had a similar AUC; among them, bipolar disorder had the lowest score, schizophrenia had the highest score, while anxiety and depression both had an AUC of approximately 0.56. Conclusion: This study is the first to conduct wide and comprehensive analyses on the use of these three scales in Chinese outpatient clinics with both traditional statistical approaches and novel machine learning methods. Our results indicated that the univariate subscale scores did not have statistical significance among our four diagnostic groups, which highlights the limit of their practical use by doctors in identifying different mental diseases in Chinese outpatient psychiatric services.