A machine learning approach to explore cognitive signatures in patients with temporo-mesial epilepsy

Innovations in Neuropsychology: Future Applications in Neurosurgical Patient Care

Over the last century, collaboration between clinical neuropsychologists and neurosurgeons has advanced the state of the science in both disciplines. These advances have provided the field of neuropsychology with many opportunities for innovation in the care of patients prior to, during, and following neurosurgical intervention. Beyond giving a general overview of how present-day advances in technology are being applied in the practice of neuropsychology within a neurological surgery department, this article outlines new developments that are currently unfolding. Improvements in remote platform, computer interface, "real-time" analytics, mobile devices, and immersive virtual reality have the capacity to increase the customization, precision, and accessibility of neuropsychological services. In doing so, such innovations have the potential to improve outcomes and ameliorate health care disparities.

Application of Improved SMOTE and XGBoost Algorithm in the Analysis of Psychological Stress Test for College Students

To address the phenomenon of serious psychological stress among college students, there are problems of high cost and subjectivity in assessing psychological stress by collecting physiological data, and this paper proposes a stress assessment method (Improved SMOTE + XGBoost) based on intelligent data collection, which divides stress levels into five levels. In the process of processing a large amount of data, there will be too little data. Therefore, this paper applies the improved SMOTE method to the data preprocessing, which can reduce the difficulty of collecting psychological stress test data while ensuring the amount of data. Firstly, we extracted features from cell phone data to generate samples, processed the samples by SMOTE, and then filtered features by XGBoost algorithm to filter features; meanwhile, we trained RF, SVM, BP, and KNN with the data before and after sampling and before and after feature screening, and the results showed that Improved SMOTE + XGBoost outperformed other methods.

Mapping of Language-and-Memory Networks in Patients With Temporal Lobe Epilepsy by Using the GE2REC Protocol

Preoperative mapping of language and declarative memory functions in temporal lobe epilepsy (TLE) patients is essential since they frequently encounter deterioration of these functions and show variable degrees of cerebral reorganization. Due to growing evidence on language and declarative memory interdependence at a neural and neuropsychological level, we propose the GE2REC protocol for interactive language-and-memory network (LMN) mapping. GE2REC consists of three inter-related tasks, sentence generation with implicit encoding (GE) and two recollection (2REC) memory tasks: recognition and recall. This protocol has previously been validated in healthy participants, and in this study, we showed that it also maps the LMN in the left TLE (N = 18). Compared to healthy controls (N = 19), left TLE (LTLE) showed widespread inter- and intra-hemispheric reorganization of the LMN through reduced activity of regions engaged in the integration and the coordination of this meta-network. We also illustrated how this protocol could be implemented in clinical practice individually by presenting two case studies of LTLE patients who underwent efficient surgery and became seizure-free but showed different cognitive outcomes. This protocol can be advantageous for clinical practice because it (a) is short and easy to perform; (b) allows brain mapping of essential cognitive functions, even at an individual level; (c) engages language-and-memory interaction allowing to evaluate the integrative processes within the LMN; (d) provides a more comprehensive assessment by including both verbal and visual modalities, as well as various language and memory processes. Based on the available postsurgical data, we presented preliminary results obtained with this protocol in LTLE patients that could potentially inform the clinical practice. This implies the necessity to further validate the potential of GE2REC for neurosurgical planning, along with two directions, guiding resection and describing LMN neuroplasticity at an individual level.

Missing links: The functional unification of language and memory (L∪M)

The field of neurocognition is currently undergoing a significant change of perspective. Traditional neurocognitive models evolved into an integrative and dynamic vision of cognitive functioning. Dynamic integration assumes an interaction between cognitive domains traditionally considered to be distinct. Language and declarative memory are regarded as separate functions supported by different neural systems. However, they also share anatomical structures (notably, the inferior frontal gyrus, the supplementary motor area, the superior and middle temporal gyrus, and the hippocampal complex) and cognitive processes (such as semantic and working memory) that merge to endorse our quintessential daily lives. We propose a new model, "L∪M" (i.e., Language/union/Memory), that considers these two functions interactively. We fractionated language and declarative memory into three fundamental dimensions or systems ("Receiver-Transmitter", "Controller-Manager" and "Transformer-Associative" Systems), that communicate reciprocally. We formalized their interactions at the brain level with a connectivity-based approach. This new taxonomy overcomes the modular view of cognitive functioning and reconciles functional specialization with plasticity in neurological disorders.

Neurobehavioural comorbidities of epilepsy: towards a network-based precision taxonomy

Cognitive and behavioural comorbidities are prevalent in childhood and adult epilepsies and impose a substantial human and economic burden. Over the past century, the classic approach to understanding the aetiology and course of these comorbidities has been through the prism of the medical taxonomy of epilepsy, including its causes, course, characteristics and syndromes. Although this 'lesion model' has long served as the organizing paradigm for the field, substantial challenges to this model have accumulated from diverse sources, including neuroimaging, neuropathology, neuropsychology and network science. Advances in patient stratification and phenotyping point towards a new taxonomy for the cognitive and behavioural comorbidities of epilepsy, which reflects the heterogeneity of their clinical presentation and raises the possibility of a precision medicine approach. As we discuss in this Review, these advances are informing the development of a revised aetiological paradigm that incorporates sophisticated neurobiological measures, genomics, comorbid disease, diversity and adversity, and resilience factors. We describe modifiable risk factors that could guide early identification, treatment and, ultimately, prevention of cognitive and broader neurobehavioural comorbidities in epilepsy and propose a road map to guide future research.

Prediction of the clinical and naming status after anterior temporal lobe resection in patients with epilepsy

By assessing the cognitive capital, neuropsychological evaluation (NPE) plays a vital role in the perioperative workup of patients with refractory focal epilepsy. In this retrospective study, we used cutting-edge statistical approaches to examine a group of 47 patients with refractory temporal lobe epilepsy (TLE), who underwent standard anterior temporal lobectomy (ATL). Our objective was to determine whether NPE may represent a robust predictor of the postoperative status, two years after surgery. Specifically, based on pre- and postsurgical neuropsychological data, we estimated the sensitivity of cognitive indicators to predict and to disentangle phenotypes associated with more or less favorable outcomes. Engel (ENG) scores were used to assess clinical outcome, and picture naming (NAM) performance to estimate naming status. Two methods were applied: (a) machine learning (ML) to explore cognitive sensitivity to postoperative outcomes; and (b) graph theory (GT) to assess network properties reflecting favorable vs. less favorable phenotypes after surgery. Specific neuropsychological indices assessing language, memory, and executive functions can globally predict outcomes. Interestingly, preoperative cognitive networks associated with poor postsurgical outcome already exhibit an atypical, highly modular and less densely interconnected configuration. We provide statistical and clinical tools to anticipate the condition after surgery and achieve a more personalized clinical management. Our results also shed light on possible mechanisms put in place for cognitive adaptation after acute injury of central nervous system in relation with surgery.

A decision tree prediction model for a short-term outcome of delirium in patients with advanced cancer receiving pharmacological interventions: A secondary analysis of a multicenter and prospective observational study (Phase-R)

Objective

There is no widely used prognostic model for delirium in patients with advanced cancer. The present study aimed to develop a decision tree prediction model for a short-term outcome.

Method

This is a secondary analysis of a multicenter and prospective observational study conducted at 9 psycho-oncology consultation services and 14 inpatient palliative care units in Japan. We used records of patients with advanced cancer receiving pharmacological interventions with a baseline Delirium Rating Scale Revised-98 (DRS-R98) severity score of ≥10. A DRS-R98 severity score of <10 on day 3 was defined as the study outcome. The dataset was randomly split into the training and test dataset. A decision tree model was developed using the training dataset and potential predictors. The area under the curve (AUC) of the receiver operating characteristic curve was measured both in 5-fold cross-validation and in the independent test dataset. Finally, the model was visualized using the whole dataset.

Results

Altogether, 668 records were included, of which 141 had a DRS-R98 severity score of <10 on day 3. The model achieved an average AUC of 0.698 in 5-fold cross-validation and 0.718 (95% confidence interval, 0.627–0.810) in the test dataset. The baseline DRS-R98 severity score (cutoff of 15), hypoxia, and dehydration were the important predictors, in this order.

Significance of results

We developed an easy-to-use prediction model for the short-term outcome of delirium in patients with advanced cancer receiving pharmacological interventions. The baseline severity of delirium and precipitating factors of delirium were important for prediction.

XGBoost Algorithm Prediction of Critical Care Outcome for Adult Patients Presenting to Emergency Department Using Initial Triage Information

Background

The emergency department (ED) triage system to classify and prioritize patients from high risk to less urgent continues to be a challenge.

Objective

This study, comprising 80,433 patients, aims to develop a machine learning algorithm prediction model of critical care outcomes for adult patients using information collected during ED triage and compare the performance with that of the baseline model using the Korean Triage and Acuity Scale (KTAS).

Methods

To predict the need for critical care, we used 13 predictors from triage information: age, gender, mode of ED arrival, the time interval between onset and ED arrival, reason of ED visit, chief complaints, systolic blood pressure, diastolic blood pressure, pulse rate, respiratory rate, body temperature, oxygen saturation, and level of consciousness. The baseline model with KTAS was developed using logistic regression, and the machine learning model with 13 variables was generated using extreme gradient boosting (XGB) and deep neural network (DNN) algorithms. The discrimination was measured by the area under the receiver operating characteristic (AUROC) curve. The ability of calibration with Hosmer–Lemeshow test and reclassification with net reclassification index were evaluated. The calibration plot and partial dependence plot were used in the analysis.

Results

The AUROC of the model with the full set of variables (0.833-0.861) was better than that of the baseline model (0.796). The XGB model of AUROC 0.861 (95% CI 0.848-0.874) showed a higher discriminative performance than the DNN model of 0.833 (95% CI 0.819-0.848). The XGB and DNN models proved better reclassification than the baseline model with a positive net reclassification index. The XGB models were well-calibrated (Hosmer-Lemeshow test; P>.05); however, the DNN showed poor calibration power (Hosmer-Lemeshow test; P<.001). We further interpreted the nonlinear association between variables and critical care prediction.

Conclusions

Our study demonstrated that the performance of the XGB model using initial information at ED triage for predicting patients in need of critical care outperformed the conventional model with KTAS.

Reconfiguration dynamics of a language-and-memory network in healthy participants and patients with temporal lobe epilepsy

Current theoretical frameworks suggest that human behaviors are based on strong and complex interactions between cognitive processes such as those underlying language and memory functions in normal and neurological populations. We were interested in assessing the dynamic cerebral substrate of such interaction between language and declarative memory, as the composite function, in healthy controls (HC, N = 19) and patients with temporal lobe epilepsy (TLE, N = 16). Our assumption was that the language and declarative memory integration is based on a language-and-memory network (LMN) that is dynamic and reconfigures according to task demands and brain status. Therefore, we explored two types of LMN dynamics, a state reconfiguration (intrinsic resting-state compared to extrinsic state assessed with a sentence recall task) and a reorganization of state reconfiguration (TLE compared to HC). The dynamics was evaluated in terms of segregation (community or module detection) and integration (connector hubs). In HC, the level of segregation was the same in both states and the mechanism of LMN state reconfiguration was shown through module change of key language and declarative memory regions with integrative roles. In TLE patients, the reorganization of LMN state reconfiguration was reflected in segregation increase and extrinsic modules that were based on shorter-distance connections. While lateral and mesial temporal regions enabled state reconfiguration in HC, these regions showed reduced flexibility in TLE. We discuss our results in a connectomic perspective and propose a dynamic model of language and declarative memory functioning. We claim that complex and interactive cognitive functions, such as language and declarative memory, should be investigated dynamically, considering the interaction between cognitive networks.