Creation of a machine learning-based prognostic prediction model for various subtypes of laryngeal cancer

Depending on the source of the blastophore, there are various subtypes of laryngeal cancer, each with a unique metastatic risk and prognosis. The forecasting of their prognosis is a pressing issue that needs to be resolved. This study comprised 5953 patients with glottic carcinoma and 4465 individuals with non-glottic type (supraglottic and subglottic). Five clinicopathological characteristics of glottic and non-glottic carcinoma were screened using univariate and multivariate regression for CoxPH (Cox proportional hazards); for other models, 10 (glottic) and 11 (non-glottic) clinicopathological characteristics were selected using least absolute shrinkage and selection operator (LASSO) regression analysis, respectively; the corresponding survival models were established; and the best model was evaluated. We discovered that RSF (Random survival forest) was a superior model for both glottic and non-glottic carcinoma, with a projected concordance index (C-index) of 0.687 for glottic and 0.657 for non-glottic, respectively. The integrated Brier score (IBS) of their 1-year, 3-year, and 5-year time points is, respectively, 0.116, 0.182, 0.195 (glottic), and 0.130, 0.215, 0.220 (non-glottic), demonstrating the model's effective correction. We represented significant variables in a Shapley Additive Explanations (SHAP) plot. The two models are then combined to predict the prognosis for two distinct individuals, which has some effectiveness in predicting prognosis. For our investigation, we established separate models for glottic carcinoma and non-glottic carcinoma that were most effective at predicting survival. RSF is used to evaluate both glottic and non-glottic cancer, and it has a considerable impact on patient prognosis and risk factor prediction.

Prediction model of no-response before the first transarterial chemoembolization for hepatocellular carcinoma: TACF score

Previous clinic models for patients with hepatocellular carcinoma (HCC) receiving transarterial chemoembolization (TACE) mainly focused on the overall survival, whereas a simple-to-use tool for predicting the response to the first TACE and the management of risk classification before TACE are lacking. Our aim was to develop a scoring system calculated manually for these patients. A total of 437 patients with hepatocellular carcinoma (HCC) who underwent TACE treatment were carefully selected for analysis. They were then randomly divided into two groups: a training group comprising 350 patients and a validation group comprising 77 patients. Furthermore, 45 HCC patients who had recently undergone TACE treatment been included in the study to validate the model's efficacy and applicability. The factors selected for the predictive model were comprehensively based on the results of the LASSO, univariate and multivariate logistic regression analyses. The discrimination, calibration ability and clinic utility of models were evaluated in both the training and validation groups. A prediction model incorporated 3 objective imaging characteristics and 2 indicators of liver function. The model showed good discrimination, with AUROCs of 0.735, 0.706 and 0.884 and in the training group and validation groups, and good calibration. The model classified the patients into three groups based on the calculated score, including low risk, median risk and high-risk groups, with rates of no response to TACE of 26.3%, 40.2% and 76.8%, respectively. We derived and validated a model for predicting the response of patients with HCC before receiving the first TACE that had adequate performance and utility. This model may be a useful and layered management tool for patients with HCC undergoing TACE.

Comparing individual and group-level simulated neurophysiological brain connectivity using the Jansen and Rit neural mass model

Computational models are often used to assess how functional connectivity (FC) patterns emerge from neuronal population dynamics and anatomical brain connections. It remains unclear whether the commonly used group-averaged data can predict individual FC patterns. The Jansen and Rit neural mass model was employed, where masses were coupled using individual structural connectivity (SC). Simulated FC was correlated to individual magnetoencephalography-derived empirical FC. FC was estimated using phase-based (phase lag index (PLI), phase locking value (PLV)), and amplitude-based (amplitude envelope correlation (AEC)) metrics to analyze their goodness of fit for individual predictions. Individual FC predictions were compared against group-averaged FC predictions, and we tested whether SC of a different participant could equally well predict participants' FC patterns. The AEC provided a better match between individually simulated and empirical FC than phase-based metrics. Correlations between simulated and empirical FC were higher using individual SC compared to group-averaged SC. Using SC from other participants resulted in similar correlations between simulated and empirical FC compared to using participants' own SC. This work underlines the added value of FC simulations using individual instead of group-averaged SC for this particular computational model and could aid in a better understanding of mechanisms underlying individual functional network trajectories.

Ancestral, Pregnancy, and Negative Early-Life Risks Shape Children's Brain (Dis)similarity to Schizophrenia

BACKGROUND

Familial, obstetric, and early-life environmental risks for schizophrenia spectrum disorder (SSD) alter normal cerebral development, leading to the formation of characteristic brain deficit patterns prior to onset of symptoms. We hypothesized that the insidious effects of these risks may increase brain similarity to adult SSD deficit patterns in prepubescent children.

METHODS

We used data collected by the Adolescent Brain Cognitive Development (ABCD) Study (N = 8940, age = 9.9 ± 0.1 years, 4307/4633 female/male), including 727 (age = 9.9 ± 0.1 years, 351/376 female/male) children with family history of SSD, to evaluate unfavorable cerebral effects of ancestral SSD history, pre/perinatal environment, and negative early-life environment. We used a regional vulnerability index to measure the alignment of a child's cerebral patterns with the adult SSD pattern derived from a large meta-analysis of case-control differences.

RESULTS

In children with a family history of SSD, the regional vulnerability index captured significantly more variance in ancestral history than traditional whole-brain and regional brain measurements. In children with and without family history of SSD, the regional vulnerability index also captured more variance associated with negative pre/perinatal environment and early-life experiences than traditional brain measurements.

CONCLUSIONS

In summary, in a cohort in which most children will not develop SSD, familial, pre/perinatal, and early developmental risks can alter brain patterns in the direction observed in adult patients with SSD. Individual similarity to adult SSD patterns may provide an early biomarker of the effects of genetic and developmental risks on the brain prior to psychotic or prodromal symptom onset.

Individual dynamic prediction of clinical endpoint from large dimensional longitudinal biomarker history: a landmark approach

Background

The individual data collected throughout patient follow-up constitute crucial information for assessing the risk of a clinical event, and eventually for adapting a therapeutic strategy. Joint models and landmark models have been proposed to compute individual dynamic predictions from repeated measures to one or two markers. However, they hardly extend to the case where the patient history includes much more repeated markers. Our objective was thus to propose a solution for the dynamic prediction of a health event that may exploit repeated measures of a possibly large number of markers.

Methods

We combined a landmark approach extended to endogenous markers history with machine learning methods adapted to survival data. Each marker trajectory is modeled using the information collected up to the landmark time, and summary variables that best capture the individual trajectories are derived. These summaries and additional covariates are then included in different prediction methods adapted to survival data, namely regularized regressions and random survival forests, to predict the event from the landmark time. We also show how predictive tools can be combined into a superlearner. The performances are evaluated by cross-validation using estimators of Brier Score and the area under the Receiver Operating Characteristic curve adapted to censored data.

Results

We demonstrate in a simulation study the benefits of machine learning survival methods over standard survival models, especially in the case of numerous and/or nonlinear relationships between the predictors and the event. We then applied the methodology in two prediction contexts: a clinical context with the prediction of death in primary biliary cholangitis, and a public health context with age-specific prediction of death in the general elderly population.

Conclusions

Our methodology, implemented in R, enables the prediction of an event using the entire longitudinal patient history, even when the number of repeated markers is large. Although introduced with mixed models for the repeated markers and methods for a single right censored time-to-event, the technique can be used with any other appropriate modeling technique for the markers and can be easily extended to competing risks setting.

Supplementary Information

The online version contains supplementary material available at (10.1186/s12874-022-01660-3).

Estimated connectivity networks outperform observed connectivity networks when classifying people with multiple sclerosis into disability groups

BACKGROUND

Multiple Sclerosis (MS), a neurodegenerative and neuroinflammatory disease, causing lesions that disrupt the brain's anatomical and physiological connectivity networks, resulting in cognitive, visual and/or motor disabilities. Advanced imaging techniques like diffusion and functional MRI allow measurement of the brain's structural connectivity (SC) and functional connectivity (FC) networks, and can enable a better understanding of how their disruptions cause disability in people with MS (pwMS). However, advanced MRI techniques are used mainly for research purposes as they are expensive, time-consuming and require high-level expertise to acquire and process. As an alternative, the Network Modification (NeMo) Tool can be used to estimate SC and FC using lesion masks derived from pwMS and a reference set of controls' connectivity networks.

OBJECTIVE

Here, we test the hypothesis that estimated SC and FC (eSC and eFC) from the NeMo Tool, based only on an individual's lesion masks, can be used to classify pwMS into disability categories just as well as SC and FC extracted from advanced MRI directly in pwMS. We also aim to find the connections most important for differentiating between no disability vs evidence of disability groups.

MATERIALS AND METHODS

One hundred pwMS (age:45.5 ± 11.4 years, 66% female, disease duration: 12.97 ± 8.07 years) were included in this study. Expanded Disability Status Scale (EDSS) was used to assess disability, 67 pwMS had no disability (EDSS < 2). Observed SC and FC were extracted from diffusion and functional MRI directly in pwMS, respectively. The NeMo Tool was used to estimate the remaining structural connectome (eSC), by removing streamlines in a reference set of tractograms that intersected the lesion mask. The NeMo Tool's eSC was used then as input to a deep neural network to estimate the corresponding FC (eFC). Logistic regression with ridge regularization was used to classify pwMS into disability categories (no disability vs evidence of disability), based on demographics/clinical information (sex, age, race, disease duration, clinical phenotype, and spinal lesion burden) and either pairwise entries or regional summaries from one of the following matrices: SC, FC, eSC, and eFC. The area under the ROC curve (AUC) was used to assess the classification performance. Both univariate statistics and parameter coefficients from the classification models were used to identify features important to differentiating between the groups.

RESULTS

The regional eSC and eFC models outperformed their observed FC and SC counterparts (p-value < 0.05), while the pairwise eSC and SC performed similarly (p = 0.10). Regional eSC and eFC models had higher AUC (0.66-0.68) than the pairwise models (0.60-0.65), with regional eFC having highest classification accuracy across all models. Ridge regression coefficients for the regional eFC and regional observed FC models were significantly correlated (Pearson's r = 0.52, p-value < 10e-7). Decreased estimated SC node strength in default mode and ventral attention networks and increased eFC node strength in visual networks was associated with evidence of disability.

DISCUSSION

Here, for the first time, we use clinically acquired lesion masks to estimate both structural and functional connectomes in patient populations to better understand brain lesion-dysfunction mapping in pwMS. Models based on the NeMo Tool's estimates of SC and FC better classified pwMS by disability level than SC and FC observed directly in the individual using advanced MRI. This work provides a viable alternative to performing high-cost, advanced MRI in patient populations, bringing the connectome one step closer to the clinic.

Development and validation of a nomogram for predicting hematoma expansion in intracerebral hemorrhage

BACKGROUND AND OBJECTIVE

To develop and validate a clinical nomogram for individualized predicting hematoma expansion (HE) in patients with Intracerebral Hemorrhage (ICH).

METHODS

A total of 1025 patients with ICH were retrospectively enrolled in the development cohort between 2010 and 2016. We identified and integrated significant factors for HE to build a nomogram. The model was subjected to validation with a separate cohort of 397 patients from the 2017-2019. The predictive accuracy and discriminative ability were measured by concordance index (C-index). The primary outcome was HE, defined as hematoma growth more than 6 mL or 33% increase in the volume.

RESULTS

A total of 1025 patients were included for univariable analysis. HE occurred in 180 patients (17.6%). The time to initial CT (≤6h vs. >6 h; p = 0.001), NIHSS score (0-4 vs. 5-14 vs. ≥15; p = 0.031), CTA spot sign (yes vs. no vs. absent; p = 0.018), hypodensities (p = 0.000), blend sign (p = 0.005), and INR (<1.2 vs. ≥1.2; p = 0.009) were identified and entered into the nomogram. The calibration curves for probability of HE showed optimal agreement between nomogram prediction and actual observation. The C-index was 0.751. The validation cohort consisted of 397 patients and HE occurred in 78 patients (19.6%). The C-index was 0.743.

CONCLUSIONS

We developed and validated a nomogram that can individually predict HE for ICH in Chinese populations. This practical prognostic nomogram may help clinicians make decision of clinical practice and design of clinical studies.

Prediction of Autism at 3 Years from Behavioural and Developmental Measures in High-Risk Infants: A Longitudinal Cross-Domain Classifier Analysis

We integrated multiple behavioural and developmental measures from multiple time-points using machine learning to improve early prediction of individual Autism Spectrum Disorder (ASD) outcome. We examined Mullen Scales of Early Learning, Vineland Adaptive Behavior Scales, and early ASD symptoms between 8 and 36 months in high-risk siblings (HR; n = 161) and low-risk controls (LR; n = 71). Longitudinally, LR and HR-Typical showed higher developmental level and functioning, and fewer ASD symptoms than HR-Atypical and HR-ASD. At 8 months, machine learning classified HR-ASD at chance level, and broader atypical development with 69.2% Area Under the Curve (AUC). At 14 months, ASD and broader atypical development were classified with approximately 71% AUC. Thus, prediction of ASD was only possible with moderate accuracy at 14 months.

Development of a prognostic score for recommended TACE candidates with hepatocellular carcinoma: A multicentre observational study

BACKGROUND & AIMS

Previous prognostic scores for transarterial chemoembolization (TACE) were mainly derived from real-world settings, which are beyond guideline recommendations. A robust model for outcome prediction and risk stratification of recommended TACE candidates is lacking. We aimed to develop an easy-to-use tool specifically for these patients.

METHODS

Between January 2010 and May 2016, 1,604 treatment-naïve patients with unresectable hepatocellular carcinoma (HCC), Child-Pugh A5-B7 and performance status 0 undergoing TACE were included from 24 tertiary centres. Patients were randomly divided into training (n = 807) and validation (n = 797) cohorts. A prognostic model was developed and subsequently validated. Predictive performance and discrimination were further evaluated and compared with other prognostic models.

RESULTS

The final presentation of the model was "linear predictor = largest tumour diameter (cm) + tumour number", which consistently outperformed other currently available models in both training and validation datasets as well as in different subgroups. The thirtieth percentile and the third quartile of the linear predictor, namely 6 and 12, were further selected as cut-off values, leading to the "six-and-twelve" score which could divide patients into 3 strata with the sum of tumour size and number ≤6, >6 but ≤12, and >12 presenting significantly different median survival of 49.1 (95% CI 43.7-59.4) months, 32.0 (95% CI 29.9-37.5) months, and 15.8 (95% CI 14.1-17.7) months, respectively.

CONCLUSIONS

The six-and-twelve score may prove an easy-to-use tool to stratify recommended TACE candidates (Barcelona Clinic Liver Cancer stage-A/B) and predict individual survival with favourable performance and discrimination. Moreover, the score could stratify these patients in clinical practice as well as help design clinical trials with comparable criteria involving these patients. Further external validation of the score is required.

LAY SUMMARY

There is currently no prognostic model specifically developed for recommended or ideal transarterial chemoembolization (TACE) candidates with hepatocellular carcinoma, despite these patients being frequently identified as the best target population in pivotal randomized controlled trials. The six-and-twelve score provides patient survival prediction, especially in ideal candidates of TACE, outperforming other currently available models in both training and validation sets, as well as different subgroups. With cut-off values of 6 and 12, the score can stratify ideal TACE candidates into 3 strata with significantly different outcomes and may shed light on risk stratification of these patients in clinical practice as well as in clinical trials.