Dimensional Neuroimaging Endophenotypes: Neurobiological Representations of Disease Heterogeneity Through Machine Learning.

Machine learning has been increasingly used to obtain individualized neuroimaging signatures for disease diagnosis, prognosis, and response to treatment in neuropsychiatric and neurodegenerative disorders. Therefore, it has contributed to a better understanding of disease heterogeneity by identifying disease subtypes with different brain phenotypic measures. In this review, we first present a systematic literature overview of studies using machine learning and multimodal magnetic resonance imaging to unravel disease heterogeneity in various neuropsychiatric and neurodegenerative disorders, including Alzheimer's disease, schizophrenia, major depressive disorder, autism spectrum disorder, and multiple sclerosis, as well as their potential in a transdiagnostic framework, where neuroanatomical and neurobiological commonalities were assessed across diagnostic boundaries. Subsequently, we summarize relevant machine learning methodologies and their clinical interpretability. We discuss the potential clinical implications of the current findings and envision future research avenues. Finally, we discuss an emerging paradigm called dimensional neuroimaging endophenotypes. Dimensional neuroimaging endophenotypes dissects the neurobiological heterogeneity of neuropsychiatric and neurodegenerative disorders into low-dimensional yet informative, quantitative brain phenotypic representations, serving as robust intermediate phenotypes (i.e., endophenotypes), presumably reflecting the interplay of underlying genetic, lifestyle, and environmental processes associated with disease etiology.

Investigating Causal Genetic Effects on Overall Survival of Glioblastoma Patients using Normalizing Flow and Structural Causal Model.

Glioblastoma (GBM) is the most common and aggressive brain tumor with short overall survival (OS) of about 15 months. Understanding the causal factors affecting the patient survival is crucial for disease prognosis and treatment planning. Although previous efforts on survival prediction using multi-omics data has yielded useful predictive models, the causation of the correlated genetic risk factors has not been addressed. Recent advances in causal deep learning models enable the study of causality from complex dataset. In this paper, we leverage the recently proposed structural causal model (SCM) with normalizing flows parameterized by deep networks to perform the counterfactual query to investigate the causal relationship between gene mutation and OS with the presence of other confounders including sex, age and radiomics features. The query amounts to the question that what the survival days will be if the gene mutation status has been changed, i.e., from mutant to non-mutant and vice versa. The trained causal model will infer the counterfactual outcome given the intervention on specific gene mutation. We apply multivariate Cox-PH model to find the genes associated with survival, and investigate the causal genetic effect by comparing the original and counterfactual survival days in a bi-directional fashion. Particularly, the following two scenarios are considered: (1) intervention on a specific gene with non-mutant status to generate the counterfactual survival days as if the gene is mutant, with which the original survival days of the subjects with that mutant gene will be compared; (2) intervention on the gene with mutant status and perform the comparison with survival days of subjects with that non-mutant gene. Our experimental results show that no causation of two correlated genes (NF1, RB1) was revealed in the cohort (n=181), while their genetic effects on OS in terms of prolonging or shortening are generally in accordance with clinical findings.

Dimensional Neuroimaging Endophenotypes: Neurobiological Representations of Disease Heterogeneity Through Machine Learning.

Machine learning has been increasingly used to obtain individualized neuroimaging signatures for disease diagnosis, prognosis, and response to treatment in neuropsychiatric and neurodegenerative disorders. Therefore, it has contributed to a better understanding of disease heterogeneity by identifying disease subtypes that present significant differences in various brain phenotypic measures. In this review, we first present a systematic literature overview of studies using machine learning and multimodal MRI to unravel disease heterogeneity in various neuropsychiatric and neurodegenerative disorders, including Alzheimer's disease, schizophrenia, major depressive disorder, autism spectrum disorder, multiple sclerosis, as well as their potential in transdiagnostic settings. Subsequently, we summarize relevant machine learning methodologies and discuss an emerging paradigm which we call dimensional neuroimaging endophenotype (DNE). DNE dissects the neurobiological heterogeneity of neuropsychiatric and neurodegenerative disorders into a low-dimensional yet informative, quantitative brain phenotypic representation, serving as a robust intermediate phenotype (i.e., endophenotype) largely reflecting underlying genetics and etiology. Finally, we discuss the potential clinical implications of the current findings and envision future research avenues.

Applications of generative adversarial networks in neuroimaging and clinical neuroscience.

Generative adversarial networks (GANs) are one powerful type of deep learning models that have been successfully utilized in numerous fields. They belong to the broader family of generative methods, which learn to generate realistic data with a probabilistic model by learning distributions from real samples. In the clinical context, GANs have shown enhanced capabilities in capturing spatially complex, nonlinear, and potentially subtle disease effects compared to traditional generative methods. This review critically appraises the existing literature on the applications of GANs in imaging studies of various neurological conditions, including Alzheimer's disease, brain tumors, brain aging, and multiple sclerosis. We provide an intuitive explanation of various GAN methods for each application and further discuss the main challenges, open questions, and promising future directions of leveraging GANs in neuroimaging. We aim to bridge the gap between advanced deep learning methods and neurology research by highlighting how GANs can be leveraged to support clinical decision making and contribute to a better understanding of the structural and functional patterns of brain diseases.

Bias in machine learning models can be significantly mitigated by careful training: Evidence from neuroimaging studies.

Despite the great promise that machine learning has offered in many fields of medicine, it has also raised concerns about potential biases and poor generalization across genders, age distributions, races and ethnicities, hospitals, and data acquisition equipment and protocols. In the current study, and in the context of three brain diseases, we provide evidence which suggests that when properly trained, machine learning models can generalize well across diverse conditions and do not necessarily suffer from bias. Specifically, by using multistudy magnetic resonance imaging consortia for diagnosing Alzheimer's disease, schizophrenia, and autism spectrum disorder, we find that well-trained models have a high area-under-the-curve (AUC) on subjects across different subgroups pertaining to attributes such as gender, age, racial groups and different clinical studies and are unbiased under multiple fairness metrics such as demographic parity difference, equalized odds difference, equal opportunity difference, etc. We find that models that incorporate multisource data from demographic, clinical, genetic factors, and cognitive scores are also unbiased. These models have a better predictive AUC across subgroups than those trained only with imaging features, but there are also situations when these additional features do not help.

Embracing the disharmony in medical imaging: A Simple and effective framework for domain adaptation.

Domain shift, the mismatch between training and testing data characteristics, causes significant degradation in the predictive performance in multi-source imaging scenarios. In medical imaging, the heterogeneity of population, scanners and acquisition protocols at different sites presents a significant domain shift challenge and has limited the widespread clinical adoption of machine learning models. Harmonization methods, which aim to learn a representation of data invariant to these differences are the prevalent tools to address domain shift, but they typically result in degradation of predictive accuracy. This paper takes a different perspective of the problem: we embrace this disharmony in data and design a simple but effective framework for tackling domain shift. The key idea, based on our theoretical arguments, is to build a pretrained classifier on the source data and adapt this model to new data. The classifier can be fine-tuned for intra-study domain adaptation. We can also tackle situations where we do not have access to ground-truth labels on target data; we show how one can use auxiliary tasks for adaptation; these tasks employ covariates such as age, gender and race which are easy to obtain but nevertheless correlated to the main task. We demonstrate substantial improvements in both intra-study domain adaptation and inter-study domain generalization on large-scale real-world 3D brain MRI datasets for classifying Alzheimer's disease and schizophrenia.

Harmonization with Flow-Based Causal Inference.

Heterogeneity in medical data, e.g., from data collected at different sites and with different protocols in a clinical study, is a fundamental hurdle for accurate prediction using machine learning models, as such models often fail to generalize well. This paper leverages a recently proposed normalizing-flow-based method to perform counterfactual inference upon a structural causal model (SCM), in order to achieve harmonization of such data. A causal model is used to model observed effects (brain magnetic resonance imaging data) that result from known confounders (site, gender and age) and exogenous noise variables. Our formulation exploits the bijection induced by flow for the purpose of harmonization. We infer the posterior of exogenous variables, intervene on observations, and draw samples from the resultant SCM to obtain counterfactuals. This approach is evaluated extensively on multiple, large, real-world medical datasets and displayed better cross-domain generalization compared to state-of-the-art algorithms. Further experiments that evaluate the quality of confounder-independent data generated by our model using regression and classification tasks are provided.

Using Atomic Force Microscopy to Measure Thickness of Passive Film on Stainless Steel Immersed in Aqueous Solution.

Employing atomic force microscopy (AFM) to measure passive film thickness on stainless steel (SS) in aqueous solution is proposed. SUS304 austenite and SUS329J4L duplex SS samples partly covered by gold were set in a minicell. To remove the original film, the SS surface but gold was etched using dilute sulfuric acid. After cleaning, open circuit potential (OCP), and distance from the sample surface to the top of the gold were measured. They were then immersed in either 1.0% NaCl; 5.0% NaCl; or aqueous solution with pH ranging from 1.0 to 10.0 and measured again. Differences between the first and subsequent measures of OCP suggested a passive film had formed in solution with pH ranging from 2.8 to 10.0. Similarly, differences among AFM measures revealed the observed film thickness increased with increase in pH and with decrease in chloride ions. Also, film thickness in water was greater than that in a vacuum. Comparison of AFM measurements of passive film on the austenite and sigma phases in sensitized SUS329J4L duplex SS revealed the film was thinner on the sigma phase containing more chromium. Taken together, these findings suggest the proposed method is applicable for measuring the thickness of passive films in aqueous solution.

Cogels of Hyaluronic Acid and Acellular Matrix for Cultivation of Adipose-Derived Stem Cells: Potential Application for Vocal Fold Tissue Engineering.

Stem cells based tissue engineering has been one of the potential promising therapies in the research on the repair of tissue diseases including the vocal fold. Decellularized extracellular matrix (DCM) as a promising scaffold has be used widely in tissue engineering; however, it remained to be an important issue in vocal fold regeneration. Here, we applied the hydrogels (hyaluronic acid [HA], HA-collagen [HA-Col], and HA-DCM) to determine the effects of hydrogel on the growth and differentiation of human adipose-derived stem cells (hADSCs) into superficial lamina propria fibroblasts. hADSCs were isolated and characterized by fluorescence-activated cell sorting. The results indicated that HA-DCM hydrogel enhanced cell proliferation and prolonged cell morphology significantly compared to HA and HA-Col hydrogel. Importantly, the differentiation of hADSCs into fibroblasts was also promoted by cogels of HA-Col and HA-DCM significantly. The differentiation of hADSCs towards superficial lamina propria fibroblasts was accelerated by the secretion of HGF, IL-8, and VEGF, the decorin and elastin expression, and the synthesis of chondroitin sulfate significantly. Therefore, the cogel of HA-DCM hydrogel was shown to be outstanding in apparent stimulation of hADSCs proliferation and differentiation to vocal fold fibroblasts through secretion of important growth factors and synthesis of extracellular matrix.

[Meta-analysis on autogenous fat injection for unilateral vocal cord paralysis].

OBJECTIVE: This study conduct a qualitative synthesis and quantitative meta-analysis of VFAFI, aimed to study whether it is a useful treatment for UVCP. METHOD: Electronic databases PubMed, YZ365. COM, WANFANG DATA, CMJD, CHKD,CNKI were searched using relevant keywords. Reported treatment outcomes were clustered into three categories,i. e. subjective, perceptual,acoustic,aerodynamic,and stroboscopic. Meta-analyses were performed on studies with numerical results using random effects model. RESULT: Five articles were identified with a total of 404 patients. All the studies reported significant improvements or decrease after VFAFI in each category of outcome measurements. Meta-analysis demonstrated a significant increase or decrease in all categories. Adverse effects include slight inflammatory reponse can resolve spontaneously within 1 month. The recurrence rate after VFAFI was high due to the self absorption. NNE and Jitter of post-operation is lower than pre-operation,there is no significantly change between the control group and experimental group; F0, Shimmer and MPT of post-operation is higher than pre-operation, there is no significantly change between the control group and experimental group. CONCLUSION: The invasiveness and morbidity of VFAFI are low and the side effects are self-limited. Meta-analyses demonstrated significant improvements or decreased from both objective and subjective measurements. Further controlled studies with longer follow-up periods and more person were included may evaluate the effectiveness of VFAFI more reliably.

Molecular cloning and functional characterization of the hepcidin gene from the convict cichlid (Amatitlania nigrofasciata) and its expression pattern in response to lipopolysaccharide challenge.

The hepcidin gene is widely expressed in many fish species and functions as an antimicrobial peptide, suggesting that it plays an important role in the innate immune system of fish. In the present study, the Amatitlania nigrofasciata hepcidin gene (AN-hepc) was cloned from the liver and its expression during an immune response was characterized. The results of quantitative PCR and RT-PCR showed that the AN-hepc transcript was most abundant in the liver. The expression of AN-hepc mRNA was significantly increased in the liver, stomach, heart, intestine, gill and muscle but was not significantly altered in the spleen, kidney, brain or skin after lipopolysaccharide challenge. The synthetic AN-hepc peptide showed a wide spectrum of antimicrobial activity in vitro toward gram-positive and gram-negative bacteria. In particular, this peptide demonstrated potent antimicrobial activity against the aquatic pathogens Vibrio alginolyticus, V. parahaemolyticus, V. vulnificus, Aeromonas hydrophila and Streptococcus agalactiae. The in vivo bacterial challenge results demonstrated that the synthetic AN-hepc peptide significantly improved the survival rate of S. agalactiae- and V. vulnificus-infected zebrafish. Taken together, these data indicate an important role for AN-hepc in the innate immunity of A. nigrofasciata and suggest its potential application in aquaculture for increasing resistance to disease.

SOX2 as prognostic factor in head and neck cancer: a systematic review and meta-analysis.

CONCLUSION: SOX2-positive head and neck cancer patients had a worse prognosis, and this was associated with common clinicopathological poor prognostic factors. OBJECTIVE: To investigate the correlation between SOX2-positive head and neck cancer and clinicopathological features and its impact on survival. METHODS: A search in PubMed and Chinese CNKI (up to 1 July 2013) was performed. Only articles in which SOX2 antigen was detected in situ localization by immunohistochemical staining were included. This meta-analysis was done using RevMan 5.2 software. Outcomes included overall survival and various clinicopathological features. RESULTS: A total of 926 gastric cancer patients from 9 studies were included. Meta-analysis showed that patients with SOX2 expression had a significantly worse 5-year overall survival compared with those with low expression (relative risk (RR) = 2.38, 95% confidence interval (CI) = 1.10-5.15, p = 0.03, random-effect). With respect to clinicopathological features, SOX2 overexpression as assessed by the immunohistochemistry method was closely correlated with tumor T stage, lymph node metastasis, and TNM stage.

[Clinical analysis of 49 cases of esthesioneuroblastoma].

OBJECTIVE: To investigate the optimal treatment and prognostic factors of esthesioneuroblastoma (ENB). METHOD: Forty-nine cases of ENB, received therapies in Chinese people liberation army general hospital from October 2004 to June 2013, were reviewed retrospectively. The therapeutical strategies and prognostic factors were analyzed using the Kaplan-Meier method. Efficacy of different therapeutical strategies was compared. RESULT: The 5-year overall survival (OS) in the group was 63.8%. The 5-year OS of Kadish stage A, B, C were 90.0%, 77.1%, 32.6%, respectively, with significant difference (χ(2) = 7.379, P = 0.03). The 5-year OS of the patients treated by surgery and radiotherapy, chemoradiotherapy, surgery with chemoradiotherapy were 68.6%, 0, 92.3%, respectively, with significant difference (χ(2) = 8.140, P = 0.02). The 5-year OS of the patients with transnasal endoscopic resection (or combined with transcranial resection) and lateral rhinotomy ( or combined with transcranial resection) were 63.7%, 50.0%, respectively, which had no significant difference (χ(2) = 3.818, P = 0.05). CONCLUSIONS: ENB has a relatively good prognosis in head and neck malignancies, the prognosis is associated with clinical stages and treatment approaches. Transnasal endoscopic surgery based multimodality is the preferable choice.

Advances in nanomedicine for head and neck cancer.

The quality of life of patients with head and neck squamous cell carcinoma (HNSCC) has been improved because of advances in surgical and radiotherapeutic techniques as well as organ-preservation methods. Despite such progresses, survival rates are dismal because of frequent recurrences, distant metastases and the development of secondary primary tumors. Nanoparticles have distinct characteristics such as a high surface/volume ratio and surface charge and size that can be easily modified. Because of such inherent features, nanoparticles are used in imaging, adjuvant radiotherapy, and drug- or gene-delivery. Thus, nanomedicine holds great promise in the diagnosis and treatment of cancer. In the present review, we summarize recent advances in nanomedicine in the diagnosis and treatment of head and neck cancer. We first review the application of inorganic nanoparticles to photo-thermal and magneto-thermal radiotherapy. We also discuss the use of organic nanoparticles in drug- or gene-delivery during chemotherapy. We then review the application of inorganic nanoparticles as radiotherapy enhancers. Finally, we address the factors that influence the biodistribution of nanoparticles in vivo.

[Clinical analysis of 182 cases esophageal foreign bodies and treatment for special cases].

OBJECTIVE: To provide clinical reference by summarizing and analyzing characteristics, therapeutic principles and methods for special cases of esophageal foreign bodies. METHOD: Statistical analysis 182 cases with esophageal foreign bodies, including 41 special cases. Surgery 165 cases, 17 cases without surgery. RESULT: One hundred and seventy-five cases were cured, 7 cases improved. Hospitalization time 0.5-50 d (average 4.5 d). CONCLUSION: It is important to learn about characteristics of esophageal foreign bodies, therapeutic principles and methods for special cases, and it's a key point to diagnosis early and remove the foreign bodies as soon as possible, and comprehensive treatment measures are really effective to process special cases.

[Comparison of the exposure when using endoscopes of different angles for microlaryngoscopic surgery].

OBJECTIVE: To observe the exposure of the vocal lesions under endoscopes of different angles and to contrast the endoscopes' influence on surgery. METHOD: Fifty patients with benign vocal lesions were selected randomly. During microlaryngoscopic surgery, the laryngoscope was placed softly and the different rigid endoscope (30 degrees, 45 degrees or 70 degrees) was connected to video system to expose the glottis and vocal folds. The operation was taken through looking at the color monitor. RESULT: The vocal fold lesions were completely examined and successfully excised in all patients, only two of them got soft palate mucosal abrasion. No one got complications. CONCLUSION: Rigid endoscopy associated with microlaryngeal surgery brings a better observation to the vocal lesions and a lower incidence rate of complicatons. It's regarded that 45 degrees rigid endoscope is better than the others, for it provides not only a good exposure to the vocal cords, anterior commissure and subglottis, but also convenience to operate. The second one is 30 degrees rigid endoscope, which can save the cost, because of the widely using in many other departments.

Hyperexcitability of inferior colliculus neurons caused by acute noise exposure.

Noise exposure is one of the most common causes of hearing loss. Recent studies found that noise exposure-induced cochlear damage may change the excitability and tonotopic organization of the central auditory system (CAS). This plasticity was suspected to be related to tinnitus and hyperacusis. However, how cochlear damage affects CAS function and causes these neurologic diseases is still not clear. CAS function is activity dependent, so we hypothesize that a restricted cochlear lesion might disrupt the balance of excitation and inhibition in the CAS and thereby affect its neural activity. To test this hypothesis, we studied the effects of narrow-band noise exposure on the firing properties of neurons in the inferior colliculus (IC), which has complex neural circuits and plays an important role in sound processing. We found that noise exposure (20 kHz, 105 dB SPL, 30 min) caused a dramatic decrease of the characteristic frequency in about two-thirds of high-frequency neurons with/without causing a significant threshold shift. The noise exposure also caused an increase in firing rate of the low-frequency neurons at suprathreshold levels, whereas it dramatically decreased the firing rate of the high-frequency neurons. Our results suggest that acute high-frequency noise exposure may increase low-frequency responsiveness by causing hyperexcitability of low-frequency neurons. The functional change of the low-frequency neurons may be related to the disruption of side-band inhibition at the noise exposure frequencies caused by cochlear damage.