OViTAD: Optimized Vision Transformer to Predict Various Stages of Alzheimer's Disease Using Resting-State fMRI and Structural MRI Data.

Advances in applied machine learning techniques for neuroimaging have encouraged scientists to implement models to diagnose brain disorders such as Alzheimer's disease at early stages. Predicting the exact stage of Alzheimer's disease is challenging; however, complex deep learning techniques can precisely manage this. While successful, these complex architectures are difficult to interrogate and computationally expensive. Therefore, using novel, simpler architectures with more efficient pattern extraction capabilities, such as transformers, is of interest to neuroscientists. This study introduced an optimized vision transformer architecture to predict the group membership by separating healthy adults, mild cognitive impairment, and Alzheimer's brains within the same age group (>75 years) using resting-state functional (rs-fMRI) and structural magnetic resonance imaging (sMRI) data aggressively preprocessed by our pipeline. Our optimized architecture, known as OViTAD is currently the sole vision transformer-based end-to-end pipeline and outperformed the existing transformer models and most state-of-the-art solutions. Our model achieved F1-scores of 97%±0.0 and 99.55%±0.39 from the testing sets for the rs-fMRI and sMRI modalities in the triple-class prediction experiments. Furthermore, our model reached these performances using 30% fewer parameters than a vanilla transformer. Furthermore, the model was robust and repeatable, producing similar estimates across three runs with random data splits (we reported the averaged evaluation metrics). Finally, to challenge the model, we observed how it handled increasing noise levels by inserting varying numbers of healthy brains into the two dementia groups. Our findings suggest that optimized vision transformers are a promising and exciting new approach for neuroimaging applications, especially for Alzheimer's disease prediction.

Electronic Rapid Fitness Assessment Identifies Factors Associated with Adverse Early Postoperative Outcomes following Radical Cystectomy.

PURPOSE: Frailty is associated with adverse outcomes following radical cystectomy. Prospective tools to identify factors affecting outcomes are needed. We describe a novel electronic rapid fitness assessment to evaluate geriatric patients undergoing radical cystectomy. MATERIALS AND METHODS: Before undergoing radical cystectomy between February 2015 and February 2018, 80 patients older than age 75 years completed the electronic rapid fitness assessment and were perioperatively comanaged by the Geriatrics Service. Physical function and cognitive function over 12 domains were evaluated and an accumulated geriatric deficit score was compiled. Hospital length of stay, discharge disposition, unplanned intensive care unit admissions, urgent care visits, readmissions, complications and deaths were assessed. RESULTS: A total of 65 patients who underwent radical cystectomy for bladder cancer without concomitant procedures completed the assessment. Median age was 80 (77, 84) years and 52 (80%) were male. A higher proportion of patients with intensive care unit admission, urgent care visit and major complications had impairments identified within electronic rapid fitness assessment domains, including Timed Up and Go. Readmission rates were similar between patients with or without deficits identified. Higher accumulated geriatric deficit score was significantly associated with intensive care unit admission (p=0.035), death within 90 days (p=0.037) and discharge to other than home (p=0.0002). CONCLUSIONS: We demonstrated the feasibility of assessing fitness in patients older than 75 years undergoing radical cystectomy using a novel electronic fitness tool. Physical limitations and overall impairment corresponded to higher intensive care unit admission rates and adverse postoperative outcomes. Larger studies in less resourced environments are required to validate these findings.

Association of Geriatric Comanagement and 90-Day Postoperative Mortality Among Patients Aged 75 Years and Older With Cancer.

Importance: Collaboration between geriatricians and surgeons in the perioperative treatment of older patients has been associated with improved outcomes in several nononcologic specialties. Similar associations may be possible among older patients with cancer. Objective: To investigate the associations of geriatric comanagement of care for older patients undergoing cancer-related surgical treatment with 90-day postoperative mortality, rate of adverse surgical events, and postoperative use of inpatient supportive care services. Design, Setting, and Participants: This retrospective cohort study assessed outcomes of patients who received geriatric comanaged care vs those who did not using multivariable logistic regression analysis, with 90-day mortality as the outcome and geriatric comanagement of care as the main variable, with adjustment for age, sex, American Society of Anesthesiology score, Memorial Sloan Kettering Frailty Index score, preoperative albumin level, operative time, and estimated blood loss. A similar model was used to assess the association of geriatric comanagement with adverse surgical events, defined as any major complication, readmission, or emergency department visit within 30 days. Patients aged 75 years and older who underwent an elective surgical procedure with a hospital stay of at least 1 day at a single tertiary-care cancer center between February 2015 and February 2018 were included. Data were analyzed from January to July 2019. Exposures: Postoperative care comanaged by the geriatrics service and surgical service (geriatric comanagement group) vs by the surgical service only (surgical service group). Main Outcomes and Measures: 90-day mortality, adverse surgical events, and use of supportive care services. Results: Of 1892 patients included, 1020 (53.9%) received geriatric comanagement of care; these patients, compared with those who received care managed by the surgery service only, were older (mean [SD] age, 81 [4] years vs 80 [4] years; P < .001), had longer operative time (mean [SD], 203 [146] minutes vs 138 [112] minutes; P < .001), and longer length of stay (median [interquartile range], 5 [3-8] days vs 4 [2-7] days; P < .001). There were no differences in the proportions of men (488 [47.8%] men vs 450 [51.6%] men; P = .11). Adverse surgical events were not significantly different between groups (odds ratio, 0.93 [95% CI, 0.73-1.18]; P = .54). However, the adjusted probability of death within 90 days after surgical treatment was 4.3% for the geriatric comanagement group vs 8.9% for the surgical service group (difference, 4.6% [95% CI, 2.3%-6.9%]; P < .001). Additionally, compared with patients who received postoperative care management from the surgery service only, a higher proportion of patients in the geriatric comanagement group received inpatient supportive care services, including physical therapy (555 patients [63.6%] vs 820 patients [80.4%]; P < .001), occupational therapy (220 patients [25.2%] vs 385 patients [37.7%]; P < .001), speech and swallow rehabilitation (42 patients [4.8%] vs 86 patients [8.4%]; P = .002), and nutrition services (637 patients [73.1%] vs 803 patients [78.7%]; P = .004). Conclusions and Relevance: This cohort study found that geriatric comanagement was associated with significantly lower 90-day postoperative mortality among older patients with cancer. These findings suggest that such patients may benefit from geriatric comanagement, which could improve their ability to survive adverse postoperative events.

Is screening for psychosocial risk factors associated with mental health care in older adults with cancer undergoing surgery?

BACKGROUND: Surgery is a notable stressor for older adults with cancer, who often are medically and psychosocially complex. The current study examined rates of preoperative psychosocial risk factors in older adults with cancer who were undergoing elective surgery and the relationship between these risk factors and the provision of mental health services during the postoperative hospitalization. METHODS: A total of 1211 patients aged ≥75 years who were referred to the geriatrics service at a comprehensive cancer center were enrolled. Patients underwent elective surgery with a length of stay of ≥3 days and were followed for at least 30 days after surgery. A comprehensive geriatric assessment was administered as part of routine preoperative care. Bivariate relationships between demographic and surgical characteristics and the preoperative comprehensive geriatric assessment and the receipt of mental health services during the postoperative hospitalization period were examined. Characteristics with bivariate relationships that were significant at the level of P < .10 were entered into a multivariable regression predicting postoperative mental health service use. RESULTS: Approximately one-fifth of the total sample (20.6%) received postoperative mental health services. In multivariable analyses, high distress (P = .007) and poor social support (P = .02) were found to be associated with a greater likelihood of the receipt of mental health services. Of those patients with high distress and poor social support, only approximately one-quarter (24.6%-25.5%) received mental health care. CONCLUSIONS: Distressed older adults and those with low levels of support preoperatively were found to be more likely to receive mental health services after surgery. Nevertheless, less than one-third of these patients received inpatient postoperative mental health care, indicating that barriers to translating screening into the provision of psychosocial services remain.

Geriatric Assessment, Not ASA Physical Status, Is Associated With 6-Month Postoperative Survival in Patients With Cancer Aged ≥75 Years.

BACKGROUND: The American Society of Anesthesiologists physical status (ASA PS) classification system is the most common method of assessing preoperative functional status. Comprehensive geriatric assessment (CGA) has been proposed as a supplementary tool for preoperative assessment of older adults. The goal of this study was to assess the correlation between ASA classification and CGA deficits among oncogeriatric patients and to determine the association of each with 6-month survival. PATIENTS AND METHODS: Oncogeriatric patients (aged ≥75 years) who underwent preoperative CGA in an outpatient geriatric clinic at a single tertiary comprehensive cancer center were identified. All patients underwent surgery, with a hospital length of stay (LOS) ≥1 day and at least 6 months of follow-up. ASA classifications were obtained from preoperative anesthesiology notes. Preoperative CGA scores ranged from 0 to 13. Six-month survival was assessed using the Social Security Death Index. RESULTS: In total, 81 of the 980 patients (8.3%) included in the study cohort died within 6 months of surgery. Most patients were classified as ASA PS III (85.4%). The mean number of CGA deficits for patients with PS II was 4.03, PS III was 5.15, and PS IV was 6.95 (P<.001). ASA classification was significantly associated with age, preoperative albumin level, hospital LOS, and 30-day intensive care unit (ICU) admissions. On multivariable analysis, 6-month mortality was associated with number of CGA deficits (odds ratio [OR], 1.14 per each unit increase in CGA score; P=.01), 30-day ICU admissions (OR, 2.77; P=.003), hospital LOS (OR, 1.03; P=.02), and preoperative albumin level (OR, 0.36; P=.004). ASA classification was not associated with 6-month mortality. CONCLUSIONS: Number of CGA deficits was strongly associated with 6-month mortality; ASA classification was not. Preoperative CGA elicits critical information that can be used to enhance the prediction of postoperative outcomes among older patients with cancer.

Geriatric co-management leads to safely performed cytoreductive surgery in older women with advanced stage ovarian cancer treated at a tertiary care cancer center.

OBJECTIVE: To assess fitness and outcomes in older women undergoing cytoreductive surgery for advanced ovarian cancer (OC). METHODS: A prospective study of OC patients referred to Geriatrics Clinic for preoperative evaluation. All completed the electronic Rapid Fitness Assessment (eRFA) and were followed by Geriatrics Service during inpatient postoperative course, co-managed by Surgical Service. Outcomes were 30-day Intensive Care Unit (ICU) admission, emergency room (ER) visit, readmission, mortality, adverse surgical events. Descriptive statistics were used. RESULTS: Forty-two women (median age 79, range 74-88), 38 with newly diagnosed advanced OC, 4 with recurrent OC, underwent cytoreductive surgery between 5/2015 and 1/2018. Preoperative age-related impairments per eRFA: high level of distress (71%), functional dependency (59%), limited social activity (59%), depression (57%), slow Time Up and Go (54%), Karnofsky Performance Score (KPS) ≤ 80 (41%), poor social support (43%), polypharmacy (35%), weight loss>10 lbs. (25%), fall history (244%), cognitive impairment (13%). Median number of comorbid conditions = 3. Among 38 newly diagnosed women, 26 (68%) had stage IIIC, 11 (29%) stage IV. Sixteen (42%) underwent primary debulking surgery, 22 (58%) neoadjuvant chemotherapy followed by interval debulking surgery. Median duration of surgery = 245.5 min (range 95-621); median hospital length of stay = 6 days (range 0-22). Optimal debulking rate = 97%, complete gross resection rate = 63%. One patient was admitted to ICU, 26% had 30-day ER visit, 10% were readmitted. Any complication, minor complication, major complication occurred in 58%, 55%, 8%, respectively. Median time from surgery to postoperative chemotherapy = 34.5 days (range 19-66). Median follow-up = 15.7 months (range 3.7-38.0), 12-month survival = 93.3%. There was no 180-day mortality. CONCLUSION: Cytoreductive surgery among older women with advanced OC and frailty can be performed safely in a tertiary care center with preoperative/postoperative geriatric and surgical co-management.

Development and Evaluation of a New Frailty Index for Older Surgical Patients With Cancer.

Importance: Frailty based on the modified Frailty Index is associated with poor postoperative outcomes. However, the index requires high levels of personnel time and effort and often has missing data. Objective: To evaluate the association of the Memorial Sloan Kettering-Frailty Index (MSK-FI) with established geriatric assessment (GA) and surgical outcomes. Design, Setting, and Participants: This cohort study included prospectively evaluated patients with cancer 75 years and older who were referred to MSK Geriatrics Service clinics for preoperative evaluation before undergoing surgery requiring hospitalization between February 2015 and September 2017. Patients were comanaged by the Geriatrics Service and Surgery Service in the postoperative period. Exposures: Impairments identified by GA and comorbid conditions retrieved from submitted International Classification of Diseases, Ninth Revision (ICD-9) and ICD-10 codes within the first 48 hours of hospitalization. Main Outcomes and Measures: The association of MSK-FI score (which included ICD-9 and ICD-10 codes) with GA impairments (based on clinical interview and examination as well as patient reports) was examined. The associations of MSK-FI score with short-term surgical outcomes (ie, frequency of complications, length of stay, 30-day surgical complications, 30-day intensive care unit admissions, and 30-day readmissions) and 1-year survival, estimated by Kaplan-Meier methods, were determined. Results: In total, 1137 patients (median [interquartile range] age, 80 [77-84] years; 583 [51.2%] women) were included in the study. A higher MSK-FI score was associated with the number of GA impairments (ρ = 0.52; bootstrapped 95% CI, 0.47-0.56). Each 1-point increase in MSK-FI score was associated with longer length of stay (0.58 d; 95% CI, 0.22-0.95; P = .002) and higher odds of intensive care unit admission (odds ratio, 1.28; 95% CI, 1.04-1.58; P = .02). Median (interquartile range) follow-up among survivors was 12.1 (5.6-19.1) months. The MSK-FI score was associated with overall mortality; 12-month risk of death was 5% for a score of 0 and approximately 20% for scores of 4 and higher (nonlinear association, P = .005). Conclusions and Relevance: In this study, the MSK-FI was associated with the previously validated GA and postoperative outcomes in older patients with cancer and may be a feasible tool for perioperative assessment of older surgical patients with cancer. Future studies should assess the association of MSK-FI score with postoperative care and outcomes of older, frail patients with cancer.

MCADNNet: Recognizing Stages of Cognitive Impairment through Efficient Convolutional fMRI and MRI Neural Network Topology Models.

Mild cognitive impairment (MCI) represents the intermediate stage between normal cerebral aging and dementia associated with Alzheimer's disease (AD). Early diagnosis of MCI and AD through artificial intelligence has captured considerable scholarly interest; researchers hope to develop therapies capable of slowing or halting these processes. We developed a state-of-the-art deep learning algorithm based on an optimized convolutional neural network (CNN) topology called MCADNNet that simultaneously recognizes MCI, AD, and normally aging brains in adults over the age of 75 years, using structural and functional magnetic resonance imaging (fMRI) data. Following highly detailed preprocessing, four-dimensional (4D) fMRI and 3D MRI were decomposed to create 2D images using a lossless transformation, which enables maximum preservation of data details. The samples were shuffled and subject-level training and testing datasets were completely independent. The optimized MCADNNet was trained and extracted invariant and hierarchical features through convolutional layers followed by multi-classification in the last layer using a softmax layer. A decision-making algorithm was also designed to stabilize the outcome of the trained models. To measure the performance of classification, the accuracy rates for various pipelines were calculated before and after applying the decision-making algorithm. Accuracy rates of 99.77% 0.36% and 97.5% 1.16% were achieved for MRI and fMRI pipelines, respectively, after applying the decision-making algorithm. In conclusion, a cutting-edge and optimized topology called MCADNNet was designed and preceded a preprocessing pipeline; this was followed by a decision-making step that yielded the highest performance achieved for simultaneous classification of the three cohorts examined.

Correlation of Venous Lactate and Time of Death in Emergency Department Patients with Noncritical Lactate Levels and Mortality from Trauma.

BACKGROUND: Serum venous lactate (LAC) levels help guide emergency department (ED) resuscitation of patients with major trauma. Critical LAC level (CLAC, ≥4.0 mmol/L) is associated with increased disease severity and higher mortality in injured patients. The characteristics of injured patients with non-CLAC (NCLAC) (<4.0 mmol/L) and death have not been previously described. OBJECTIVES: (1) To describe the characteristics of patients with venous NCLAC and death from trauma. (2) To assess the correlation of venous NCLAC with time of death. METHODS: A retrospective cohort study at an urban teaching hospital between 9/2011 and 8/2014. Inclusion: All trauma patients (all ages) who presented to the ED with any injury and met all criteria: (1) Venous LAC drawn at the time of arrival that resulted in an NCLAC level; (2) were admitted to the hospital; (3) died during their hospitalization. Exclusion: CLAC. Outcome: Correlation of NCLAC and time of death. Data were extracted from an electronic medical record by trained data abstractors using a standardized protocol. Cross-checks were performed on 10% of data entries and inter-observer agreement was calculated. Data were explored using descriptive statistics and Kaplan-Meier curves were created to define survival estimates. Data are presented as percentages with 95% confidence interval (CI) for proportions and medians with quartiles for continuous variables. Kaplan-Meier curves with differences in time to events based on LAC are used to analyze the data. RESULTS: A total of 60 patients met the inclusion criteria. The median age was 52 years (quartiles: 30, 75) and 73% were male (age range 2-92). The median LAC in the overall cohort was 1.9 mmol/L (quartiles: 1.5, 2.1). Sixteen patients (27%) died during the first 24 h with 5 (31%) due to intracranial hemorrhage. The median survival time was 5.6 days (134.4 h) (95% CI: 2.3-12.6). CONCLUSIONS: In trauma patients with NCLAC who died during the index hospitalization, the median survival time was 5.6 days, approximately one-third of patients died within the first 24 h. These findings indicate that relying on a triage NCLAC level alone may result in underestimating injury severity and subsequent morbidity and mortality.

Task-linked Diurnal Brain Network Reorganization in Older Adults: A Graph Theoretical Approach.

Testing older adults in the morning generally improves behavioral performance relative to afternoon testing. Morning testing is also associated with brain activity similar to that of young adults. Here, we used graph theory to explore how time of day (TOD) affects the organization of brain networks in older adults across rest and task states. We used nodes from the automated anatomical labeling atlas to construct participant-specific correlation matrices of fMRI data obtained during 1-back tasks with interference and rest. We computed pairwise group differences for key graph metrics, including small-worldness and modularity. We found that older adults tested in the morning and young adults did not differ on any graph metric. Both of these groups differed from older adults tested in the afternoon during the tasks-but not rest. Specifically, the latter group had lower modularity and small-worldness (indices of more efficient network organization). Across all groups, higher modularity and small-worldness strongly correlated with reduced distractibility on an implicit priming task. Increasingly, TOD is seen as important for interpreting and reproducing neuroimaging results. Our study emphasizes how TOD affects brain network organization and executive control in older adults.

Age differences in the functional interactions among the default, frontoparietal control, and dorsal attention networks.

Older adults typically show weaker functional connectivity (FC) within brain networks compared with young adults, but stronger functional connections between networks. Our primary aim here was to use a graph theoretical approach to identify age differences in the FC of 3 networks-default mode network (DMN), dorsal attention network, and frontoparietal control (FPC)-during rest and task conditions and test the hypothesis that age differences in the FPC would influence age differences in the other networks, consistent with its role as a cognitive "switch." At rest, older adults showed lower clustering values compared with the young, and both groups showed more between-network connections involving the FPC than the other 2 networks, but this difference was greater in the older adults. Connectivity within the DMN was reduced in older compared with younger adults. Consistent with our hypothesis, between-network connections of the FPC at rest predicted the age-related reduction in connectivity within the DMN. There was no age difference in within-network FC during the task (after removing the specific task effect), but between-network connections were greater in older adults than in young adults for the FPC and dorsal attention network. In addition, age reductions were found in almost all the graph metrics during the task condition, including clustering and modularity. Finally, age differences in between-network connectivity of the FPC during both rest and task predicted cognitive performance. These findings provide additional evidence of less within-network but greater between-network FC in older adults during rest but also show that these age differences can be altered by the residual influence of task demands on background connectivity. Our results also support a role for the FPC as the regulator of other brain networks in the service of cognition. Critically, the link between age differences in inter-network connections of the FPC and DMN connectivity, and the link between FPC connectivity and performance, support the hypothesis that FC of the FPC influences the expression of age differences in other networks, as well as differences in cognitive function.

The Associative Memory Deficit in Aging Is Related to Reduced Selectivity of Brain Activity during Encoding.

Human aging is characterized by reductions in the ability to remember associations between items, despite intact memory for single items. Older adults also show less selectivity in task-related brain activity, such that patterns of activation become less distinct across multiple experimental tasks. This reduced selectivity or dedifferentiation has been found for episodic memory, which is often reduced in older adults, but not for semantic memory, which is maintained with age. We used fMRI to investigate whether there is a specific reduction in selectivity of brain activity during associative encoding in older adults, but not during item encoding, and whether this reduction predicts associative memory performance. Healthy young and older adults were scanned while performing an incidental encoding task for pictures of objects and houses under item or associative instructions. An old/new recognition test was administered outside the scanner. We used agnostic canonical variates analysis and split-half resampling to detect whole-brain patterns of activation that predicted item versus associative encoding for stimuli that were later correctly recognized. Older adults had poorer memory for associations than did younger adults, whereas item memory was comparable across groups. Associative encoding trials, but not item encoding trials, were predicted less successfully in older compared with young adults, indicating less distinct patterns of associative-related activity in the older group. Importantly, higher probability of predicting associative encoding trials was related to better associative memory after accounting for age and performance on a battery of neuropsychological tests. These results provide evidence that neural distinctiveness at encoding supports associative memory and that a specific reduction of selectivity in neural recruitment underlies age differences in associative memory.

The prognostic role of non-critical lactate levels for in-hospital survival time among ED patients with sepsis.

OBJECTIVE: This study describes emergency department (ED) sepsis patients with non-critical serum venous lactate (LAC) levels (LAC <4.0 mmol/L) who suffered in-hospital mortality and examines LAC in relation to survival times. METHODS: An ED based retrospective cohort study accrued September 2010 to August 2014. Inclusion criteria were ED admission, LAC sampling, >2 systemic inflammatory response syndrome criteria with an infectious source (sepsis), and in-hospital mortality. Kaplan-Meier curves were used for survival estimates. An a priori sub-group analysis for patients with repeat LAC within 6 hours of initial sampling was undertaken. The primary outcome was time to in-hospital death evaluated using rank-sum tests and regression models. RESULTS: One hundred ninety-seven patients met inclusion criteria. Pulmonary infections were the most common (44%) and median LAC was 1.9 mmol/L (1.5, 2.5). Thirteen patients (7%) died within 24 hours and 79% by ≤28 days. Median survival was 11 days (95% CI, 8.0-13). Sixty-two patients had repeat LAC sampling with 14 (23%) and 48 (77%) having decreasing increasing levels, respectively. No significant differences were observed in treatment requirements between the LAC subgroups. Among patients with decreasing LAC, median survival was 24 days (95% CI, 5-32). For patients with increasing LAC median survival was significantly shorter (7 days; 95% CI, 4-11, P = .04). Patients with increasing LAC had a non-significant trend toward reduced survival (HR = 1.6 95% CI, 0.90-3.0, P = .10). CONCLUSIONS: In septic ED patients experiencing in-hospital death, non-critical serum venous lactate may be utilized as a risk-stratifying tool for early mortality, while increasing LAC levels may identify those in danger of more rapid deterioration.