RESUMEN
BACKGROUND: Oral potentially malignant disorders (OPMDs) are associated with an increased risk of cancer of the oral cavity including the tongue. The early detection of oral cavity cancers and OPMDs is critical for reducing cancer-specific morbidity and mortality. Recently, there have been studies to apply the rapidly advancing technology of deep learning for diagnosing oral cavity cancer and OPMDs. However, several challenging issues such as class imbalance must be resolved to effectively train a deep learning model for medical imaging classification tasks. The aim of this study is to evaluate a new technique of artificial intelligence to improve the classification performance in an imbalanced tongue lesion dataset. METHODS: A total of 1,810 tongue images were used for the classification. The class-imbalanced dataset consisted of 372 instances of cancer, 141 instances of OPMDs, and 1,297 instances of noncancerous lesions. The EfficientNet model was used as the feature extraction model for classification. Mosaic data augmentation, soft labeling, and curriculum learning (CL) were employed to improve the classification performance of the convolutional neural network. RESULTS: Utilizing a mosaic-augmented dataset in conjunction with CL, the final model achieved an accuracy rate of 0.9444, surpassing conventional oversampling and weight balancing methods. The relative precision improvement rate for the minority class OPMD was 21.2%, while the relative [Formula: see text] score improvement rate of OPMD was 4.9%. CONCLUSIONS: The present study demonstrates that the integration of mosaic-based soft labeling and curriculum learning improves the classification performance of tongue lesions compared to previous methods, establishing a foundation for future research on effectively learning from imbalanced data.
Asunto(s)
Aprendizaje Profundo , Neoplasias de la Boca , Humanos , Inteligencia Artificial , Curriculum , LenguaRESUMEN
Serotonergic neurotransmission has been associated with aggression in several psychiatric disorders. Human aggression is a continuum of traits, ranging from normal to pathological phenomena. However, the individual differences in serotonergic neurotransmission and their relationships with aggression traits in healthy individuals remain unclear. In this study, we explored the relationship between 5-HT2A receptor availability in vivo and aggression traits in healthy participants. Thirty-three healthy participants underwent 3-Tesla magnetic resonance imaging and positron emission tomography (PET) with [11C]MDL100907, a selective radioligand for 5-HT2A receptors. To quantify 5-HT2A receptor availability, the binding potential (BPND) was derived using the basis function implementation of the simplified reference tissue model, with the cerebellum as the reference region. The participants' aggression levels were assessed using the Buss-Perry Aggression Questionnaire. The voxel-based correlation analysis with age and sex as covariates revealed that the total aggression score was significantly positively correlated with [11C]MDL100907 BPND in the right middle temporal gyrus (MTG) pole, left fusiform gyrus (FUSI), right parahippocampal gyrus, and right hippocampus. The physical aggression subscale score had significant positive correlations with [11C]MDL100907 BPND in the left olfactory cortex, left orbital superior frontal gyrus (SFG), right anterior cingulate and paracingulate gyri, left orbitomedial SFG, left gyrus rectus, left MTG, left inferior temporal gyrus, and left angular gyrus. The verbal aggression subscale score showed significant positive correlations with [11C]MDL100907 BPND in the bilateral SFG, right medial SFG, left FUSI, and right MTG pole. Overall, our findings suggest the possibility of positive correlations between aggression traits and in vivo 5-HT2A receptor availability in healthy individuals. Future research should incorporate multimodal neuroimaging to investigate the downstream effects of 5-HT2A receptor-mediated signaling and integrate molecular and systems-level information in relation to aggression traits.
Asunto(s)
Receptor de Serotonina 5-HT2A , Serotonina , Humanos , Agresión/fisiología , Encéfalo/diagnóstico por imagen , Encéfalo/patología , Radioisótopos de Carbono , Imagen por Resonancia Magnética , Tomografía de Emisión de Positrones/métodosRESUMEN
Genetic research has implicated dopamine neurotransmission in the expression of the self-transcendence trait in humans. However, molecular imaging of dopaminergic markers is undocumented in relation to this personality trait. In this multimodal imaging study, we first investigated the relationship between the self-transcendence trait and in vivo dopamine D2/3 receptor availability using [18 F]fallypride positron emission tomography (PET). We next conducted seed-based functional connectivity analyses using resting-state functional magnetic resonance imaging (rs-fMRI) data with regions derived from the PET analysis as seeds to explore the functional significance of D2/3 receptor availability foci associated with the self-transcendence trait. Twenty-one healthy subjects underwent high-resolution PET with [18 F]fallypride and a subset of 18 subjects also completed 3-Tesla rs-fMRI. The Temperament and Character Inventory was used to measure the self-transcendence trait. A voxel-based whole brain analysis revealed that the [18 F]fallypride binding potential (BPND ) within the cluster of the left insula was significantly positively correlated with self-transcendence trait scores. A region-of-interest analysis also showed a significant positive correlation between self-transcendence and [18 F]fallypride BPND in the left insula. The exploratory [18 F]fallypride BPND seed-based rs-fMRI analysis showed that the functional connectivity from the left insula seed to the prefrontal cortices (including the inferior frontal region) was negatively associated with self-transcendence trait scores. The results of the present study suggest that D2/3 receptor-mediated neurotransmission in the left insula may constitute a significant neurobiological factor in the self-transcendence trait. The negative associations between BPND seed-based functional connectivity and self-transcendence trait scores may suggest reduced prefrontal control in this personality trait.
Asunto(s)
Encéfalo/diagnóstico por imagen , Receptores de Dopamina D2/metabolismo , Receptores de Dopamina D3/metabolismo , Autoimagen , Adulto , Benzamidas , Encéfalo/metabolismo , Femenino , Radioisótopos de Flúor , Humanos , Imagen por Resonancia Magnética , Masculino , Persona de Mediana Edad , Tomografía de Emisión de Positrones , Adulto JovenRESUMEN
AIM: The aim of this study was to assess neuropsychological correlations with the T2* relaxation time (T2*-RT) of hippocampal subregions in adolescents using ultra-high-field (UHF) 7.0-T magnetic resonance imaging (MRI). METHODS: We assessed the T2*-RT of hippocampal subregions in 31 healthy 11th- or 12th-grade high school students using an UHF 7.0-T MRI system. T2*-RT of the cornu ammonis (CA) 1, CA2, CA3, and CA4 subregions and the subiculum were calculated for both the left and right hippocampus. Seven subtests of the Cambridge Neuropsychological Test Automated Battery were administered to the subjects to assess visuospatial memory. RESULTS: Poor performances in delayed recall in the pattern-recognition test were significantly correlated with longer T2*-RT in the bilateral subiculum (right, r = -0.480, P = 0.006; left, r = -0.648, P < 0.001) and the left CA2 (r = -0.480, P = 0.006). CONCLUSION: This study showed that longer T2*-RT in the subiculum were associated with poorer performances in delayed recall in the visual memory tasks. This finding suggests that the subiculum might play a predominant role in delayed recall in adolescents.
Asunto(s)
Desarrollo del Adolescente/fisiología , Hipocampo/diagnóstico por imagen , Imagen por Resonancia Magnética , Recuerdo Mental/fisiología , Neuroimagen , Reconocimiento Visual de Modelos/fisiología , Reconocimiento en Psicología/fisiología , Adolescente , Femenino , Humanos , Imagen por Resonancia Magnética/instrumentación , Masculino , Neuroimagen/instrumentaciónRESUMEN
In addition to probing regional differences in receptor availability, molecular positron emission tomography (PET) is proving useful for investigating perturbations in neurotransmitter networks using interregional correlation analyses. In a multi-modal imaging study, we examined interregional correlations of dopamine D2/3 receptor availability between striatal and extrastriatal regions using [18 F]fallypride high-resolution PET in 11 patients with schizophrenia receiving low-dose maintenance atypical antipsychotics and 14 healthy control subjects, and investigated resting-state functional connectivity in the same subjects using seed-based functional magnetic resonance imaging (fMRI) analysis. In the healthy control group, there were no significant correlations of [18 F]fallypride binding potential (BPND ) between striatal regions and any cortical areas, whereas the patient group showed significant and widespread inter-correlations. Correlations between BPND in striatum, amygdala and insula with cortex were significantly higher in patients than in controls. In seed-based resting-state fMRI analysis, the healthy controls revealed positive and negative functional connectivity patterns, while patients exhibited a pattern of exclusively positive connectivity. Functional connectivity was significantly higher between striatal regions and extrastriatal areas including cortical regions in patients compared to controls. In this first such report, molecular and functional connectivity between striatal and extrastriatal regions was primarily characterized by increased interregional relationships in treated patients with schizophrenia. The results suggest that the spatial organization of D2/3 receptor availability and related functional connectivity are significantly perturbed in stable outpatients on maintenance antipsychotics. Future studies should include antipsychotic-naïve patients to determine if these relationships are illness-related characteristics, or arising due to chronic antipsychotic treatment.
Asunto(s)
Conectoma , Cuerpo Estriado/diagnóstico por imagen , Imagen por Resonancia Magnética , Tomografía de Emisión de Positrones , Esquizofrenia/diagnóstico por imagen , Adulto , Antipsicóticos/uso terapéutico , Benzamidas , Cuerpo Estriado/fisiopatología , Femenino , Humanos , Masculino , Persona de Mediana Edad , Pirrolidinas , Radiofármacos , Esquizofrenia/tratamiento farmacológico , Esquizofrenia/fisiopatologíaRESUMEN
Background: Previous studies have reported functional and structural abnormalities in the thalamus and the pars triangularis of the inferior frontal gyrus in patients with insomnia disorder. However, no studies have been conducted on the white-matter tracts between these 2 brain regions. We aimed to compare the white-matter integrity and structure of the left thalamuspars triangularis tracts between patients with insomnia and controls, and to characterize the relationship between white-matter integrity and clinical features in patients with insomnia. Methods: In total, 22 participants with insomnia disorder and 27 controls underwent overnight polysomnography and brain magnetic resonance imaging, and then completed self-report clinical questionnaires and neurocognitive tests for spatial planning. Structural and diffusion measures such as fractional anisotropy, axial diffusivity, radial diffusivity and trace were analyzed in group comparison and correlation analyses. Results: The insomnia group showed significantly lower fractional anisotropy (F = 8.647, p = 0.02) and axial diffusivity (F = 5.895, p = 0.038) in the left thalamuspars triangularis tracts than controls. In patients with insomnia, fractional anisotropy in the tracts was correlated with the results of the Stockings of Cambridge test (r = 0.451, p = 0.034), and radial diffusivity was correlated with Epworth Sleepiness Scale score (r = 0.437, p = 0.042). Limitations: Limitations included analyses of limited brain regions and the cross-sectional design. Conclusion: The insomnia group showed decreased integrity in the left thalamuspars triangularis tracts, and integrity was correlated with cognition and daytime sleepiness. These results may imply that insomnia is characterized by disintegration of the white-matter tract between the left thalamus and inferior frontal gyrus.
Asunto(s)
Corteza Prefrontal/diagnóstico por imagen , Trastornos del Inicio y del Mantenimiento del Sueño/diagnóstico por imagen , Tálamo/diagnóstico por imagen , Sustancia Blanca/diagnóstico por imagen , Adolescente , Adulto , Anisotropía , Estudios Transversales , Imagen de Difusión por Resonancia Magnética , Femenino , Humanos , Masculino , Persona de Mediana Edad , Pruebas Neuropsicológicas , Polisomnografía , Autoinforme , Trastornos del Inicio y del Mantenimiento del Sueño/psicología , Tractos Espinotalámicos/diagnóstico por imagen , Adulto JovenRESUMEN
The purpose of this study was to investigate the relationship between specific symptom severity and D2/3 receptor availability in extrastriatal regions in outpatients with schizophrenia to shed light on the role of extrastriatal dopaminergic neurotransmission in the pathophysiology of symptoms of schizophrenia. Sixteen schizophrenia patients receiving relatively low-dose maintenance atypical antipsychotics and seventeen healthy controls underwent 3-Tesla magnetic resonance imaging and high-resolution positron emission tomography with [18F]fallypride. For D2/3 receptor availability, the binding potential with respect to non-displaceable compartment (BPND) was derived using the simplified reference tissue model. The BPND values were lower in patients on antipsychotic treatment than in controls across all regions with large effect sizes (1.03-1.42). The regions with the largest effect size were the substantia nigra, amygdala, and insula. Symptoms of schizophrenia were assessed using a five-factor model of the Positive and Negative Syndrome Scale (PANSS). The region of interest-based analysis showed that PANSS excitement factor score had a significant positive correlation with the [18F]fallypride BPND in the insula. The equivalent dose of antipsychotics was not significantly correlated with PANSS factor scores or regional BPND values. The voxel-based analysis also revealed a significant positive association between the PANSS excitement factor and the [18F]fallypride BPND in the insula. The present study revealed a significant association between excitement symptom severity and D2/3 receptor availability in the insula in schizophrenia, suggesting a possible important role of D2/3 receptor-mediated neurotransmission in the insula and related limbic system in the pathophysiology of this specific symptom cluster.
Asunto(s)
Benzamidas , Corteza Cerebral/metabolismo , Antagonistas de los Receptores de Dopamina D2 , Tomografía de Emisión de Positrones/métodos , Receptores de Dopamina D2/metabolismo , Esquizofrenia/metabolismo , Esquizofrenia/fisiopatología , Adulto , Antipsicóticos/farmacología , Corteza Cerebral/diagnóstico por imagen , Femenino , Radioisótopos de Flúor , Humanos , Masculino , Persona de Mediana Edad , Receptores de Dopamina D3/metabolismo , Esquizofrenia/diagnóstico por imagen , Esquizofrenia/tratamiento farmacológico , Adulto JovenRESUMEN
PURPOSE: This study used functional magnetic resonance imaging (fMRI) to investigate differences in the functional brain activation of patients with insomnia disorder (n = 21, mean age = 36.6) and of good sleepers (n = 26, mean age = 33.2) without other comorbidities or structural brain abnormalities during a working memory task. METHODS: All participants completed a clinical questionnaire, were subjected to portable polysomnography (PSG), and performed the working memory task during an fMRI scan. The subjects who were suspected of major sleep disorder and comorbid psychiatric disorders except insomnia disorder were excluded. To compare the brain activation on working memory from the insomnia group with those from the good-sleeper group, a two-sample t test was performed. Statistical significance was determined using 3DClustSim with the updated algorithm to obtain a reasonable cluster size and p value for each analysis. RESULTS: We observed higher levels of brain activation in the right lateral inferior frontal cortex and the right superior temporal pole in the insomnia group compared to good sleepers (cluster-based multiple comparison correction, p < 0.001, k = 34 @ α = 0.01). CONCLUSION: Thus, patients with insomnia disorder showed increased brain activation during working memory relative to good sleepers, and this may be indicative of compensatory brain activation to maintain cognitive performance in patients with insomnia disorder without other comorbidities.
Asunto(s)
Imagen por Resonancia Magnética , Memoria a Corto Plazo/fisiología , Neuroimagen , Trastornos del Inicio y del Mantenimiento del Sueño/fisiopatología , Adulto , Femenino , Humanos , MasculinoRESUMEN
PURPOSE: To propose a new Extended Monopole antenna Array with individual Shields (EMAS) coil that improves the B1 field coverage and uniformity along the z-direction. METHODS: To increase the spatial coverage of Monopole antenna Array (MA) coil, each monopole antenna was shielded and extended in length. Performance of this new coil, which is referred to as EMAS coil, was compared with the original MA coil and an Extended Monopole antenna Array coil with no shield (EMA). For comparison, flip angle, signal-to-noise ratio (SNR), and receive sensitivity maps were measured at multiple regions of interest (ROIs) in the brain. RESULTS: The EMAS coil demonstrated substantially larger flip angle and receive sensitivity than the MA and EMA coils in the inferior aspect of the brain. In the brainstem ROI, for example, the flip angle in the EMAS coil was increased by 45.5% (or 60.0%) and the receive sensitivity was increased by 26.9% (or 14.9%), resulting in an SNR gain of 84.8% (or 76.3%) when compared with the MA coil (or EMA). CONCLUSION: The EMAS coil provided 25.7% (or 24.4%) more uniform B1+ field distribution compared with the MA (or EMA) coil in sagittal. The EMAS coil successfully extended the imaging volume in lower part of the brain. Magn Reson Med 75:2566-2572, 2016. © 2015 Wiley Periodicals, Inc.
Asunto(s)
Encéfalo/diagnóstico por imagen , Procesamiento de Imagen Asistido por Computador/métodos , Imagen por Resonancia Magnética/instrumentación , Imagen por Resonancia Magnética/métodos , Diseño de Equipo , Humanos , Relación Señal-RuidoRESUMEN
OBJECTIVE: This study investigated regional gray matter (GM) volume differences between suicide attempters and suicide non-attempters with major depressive disorder (MDD) and their relationship with psychological risk factors for suicidality. METHODS: MDD patients with and without a suicide attempt history (n=19 in each group) participated. The Hamilton Depression Rating Scale, Clinical Global Impression (severity subscale), Scale for Suicide Ideation (SSI), Risk-Rescue Rating (RRR), Beck Hopelessness Scale (BHS), Barrett Impulsivity Scale, Eysenck Personality Questionnaire, and Ways of Coping Checklist (WCCL) were administered. T1-weighted structural magnetic resonance imaging scans were acquired to evaluate changes in GM volume. Voxel-based morphometry was performed using the SPM 8 software package. Two-sample t-tests were used during second-level group comparison analysis; partial correlation analysis controlling for gender and age identified associations between regional GM volume and psychological measures. RESULTS: Suicide attempters exhibited significantly decreased GM volume in the left angular gyrus (p<0.001, uncorrected) and right cerebellum (p<0.001, uncorrected). GM volume in the left angular gyrus was inversely correlated with BHS scores (r=-0.55, p<0.01) and positively correlated with the Seeking Social Support subscale of the WCCL (r=0.43, p<0.01). CONCLUSION: These findings provide evidence of a neural basis of suicidal behaviors in MDD. In particular, reduced GM volume in the left angular gyrus may be a neurobiological marker of suicidality in depressed patients.
Asunto(s)
Trastorno Depresivo Mayor/fisiopatología , Sustancia Gris/patología , Lóbulo Parietal/patología , Intento de Suicidio/estadística & datos numéricos , Suicidio/estadística & datos numéricos , Adulto , Cerebelo , Trastorno Depresivo Mayor/psicología , Femenino , Humanos , Imagen por Resonancia Magnética , Masculino , Persona de Mediana Edad , Tamaño de los Órganos , Escalas de Valoración Psiquiátrica , Ideación Suicida , Suicidio/psicología , Intento de Suicidio/psicologíaRESUMEN
A 16-year-old, castrated, male English cocker spaniel dog was presented due to generalized alopecia. Routine clinical pathology, endocrine and abdominal ultrasonography results were consistent with a diagnosis of pituitary-dependent hyperadrenocorticism. The adenohypophyseal lesion was clearly visualized on both 3 T and 7 T magnetic resonance imaging (MRI) of the pituitary gland. Although biochemical and MRI findings were consistent with a functional pituitary microtumor, a pituitary lesion was not detected using (18)F-fluorodeoxyglucose positron emission tomography (FDG-PET). This report firstly describes the application of high-resolution FDG-PET to a spontaneous pituitary microtumor in a dog.
RESUMEN
PURPOSE: To measure the activity of individual raphe nuclei with fluorine 18 fluorodeoxyglucose (FDG) and carbon 11 ((11)C) 3-amino-4-(2-dimethylaminomethylphenylthio) benzonitrile (DASB) imaging using a brain positron emission tomography(PET)/magnetic resonance (MR) imaging fusion system. MATERIALS AND METHODS: The study was approved by the Institutional Review Board of Gil Medical Center, and all volunteers provided written informed consent. FDG PET, (11)C-DASB PET, and T2*-weighted MR images from seven healthy volunteers were acquired by using a PET/MR imaging fusion system. The standard uptake value ratio (SUVR) of FDG (FDG-SUVR) and nondisplaceable binding potential (BPnd) of (11)C-DASB (DASB-BPnd) were determined for each raphe nucleus. A Pearson correlation analysis was performed to show the correlation between FDG-SUVR and DASB-BPnd for the raphe nuclei. RESULTS: Each raphe nucleus could be distinguished in both FDG (identifiability ratio, 0.86; κ = 0.77) and (11)C-DASB (identifiability ratio, 0.89; κ = 0.72) images. The mean values of DASB-BPnd for each raphe nucleus from dorsal to caudal direction were 6.08 (raphe nucleus 1), 5.93 (raphe nucleus 2), 3.86 (raphe nucleus 3), 3.18 (raphe nucleus 4), and 2.74 (raphe nucleus 5); the mean FDG-SUVR values were 1.00 (raphe nucleus 1), 1.00 (raphe nucleus 2), 0.87 (raphe nucleus 3), 0.94 (raphe nucleus 4), and 0.90 (raphe nucleus 5). FDG-SUVR and DASB-BPnd for the raphe nuclei were significantly correlated (r = 0.506, P = .002). CONCLUSION: Serotonergic activity, both glucose metabolism and transporter binding potential of raphe nuclei, were measured with a brain-dedicated PET/MR imaging system and showed a significant correlation.
Asunto(s)
Imagen por Resonancia Magnética/métodos , Imagen Multimodal , Tomografía de Emisión de Positrones/métodos , Núcleos del Rafe/anatomía & histología , Adulto , Bencilaminas/administración & dosificación , Radioisótopos de Carbono/administración & dosificación , Femenino , Fluorodesoxiglucosa F18/administración & dosificación , Humanos , Masculino , Radiofármacos/administración & dosificaciónRESUMEN
A 45-yr-old female patient was admitted with one-month history of headache and progressive left hemiparesis. Brain magnetic resonance imaging (MRI) demonstrated a mass lesion in her right frontal lobe. Her brain tumor was confirmed as a small cell glioblastoma. Her follow-up brain MRI, taken at 8 months after her initial surgery demonstrated tumor recurrence in the right frontal lobe. Contrast-enhanced 7.0T brain magnetic resonance imaging (MRI) was safely performed before surgery and at the time of recurrence. Compared with 1.5T and 3.0T brain MRI, 7.0T MRI showed sharpened images of the brain tumor contexture with detailed anatomical information. The fused images of 7.0T and 1.5T brain MRI taken at the time of recurrence demonstrated no significant discrepancy in the positions of the anterior and the posterior commissures. It is suggested that 7.0T MRI can be safely utilized for better images of the maligant gliomas before and after surgery.
Asunto(s)
Neoplasias Encefálicas/diagnóstico por imagen , Glioblastoma/diagnóstico por imagen , Imagen por Resonancia Magnética , Neoplasias Encefálicas/patología , Neoplasias Encefálicas/cirugía , Receptores ErbB/genética , Femenino , Lóbulo Frontal/diagnóstico por imagen , Glioblastoma/patología , Glioblastoma/cirugía , Humanos , Hibridación Fluorescente in Situ , Persona de Mediana Edad , Recurrencia Local de Neoplasia , RadiografíaRESUMEN
OBJECTIVE: Human impulsivity is a complex multidimensional construct encompassing cognitive, emotional, and behavioural aspects. Previous animal studies have suggested that striatal dopamine receptors play a critical role in impulsivity. In this study, we investigated the relationship between self-reported impulsiveness and dopamine D(2/3) receptor availability in striatal subdivisions in healthy subjects using high-resolution positron emission tomography (PET) with [11C]raclopride. METHODS: Twenty-one participants completed 3-T magnetic resonance imaging and high-resolution PET scans with [11C]raclopride. The trait of impulsiveness was measured using the Barratt Impulsiveness Scale (BIS-11). Partial correlation analysis was performed between BIS-11 scores and D(2/3) receptor availability in striatal subregions, controlling for the confounding effects of temperament characteristics that are conceptually or empirically related to dopamine, which were measured by the Temperament and Character Inventory. RESULTS: The analysis revealed that the non-planning (p = 0.004) and attentional (p = 0.007) impulsiveness subscale scores on the BIS-11 had significant positive correlations with D(2/3) receptor availability in the pre-commissural dorsal caudate. There was a tendency towards positive correlation between non-planning impulsiveness score and D(2/3) receptor availability in the post-commissural caudate. CONCLUSION: These results suggest that cognitive subtrait of impulsivity is associated with D(2/3) receptor availability in the associative striatum that plays a critical role in cognitive processes involving attention to detail, judgement of alternative outcomes, and inhibitory control.
Asunto(s)
Cuerpo Estriado/metabolismo , Conducta Impulsiva/fisiología , Receptores de Dopamina D2/metabolismo , Receptores de Dopamina D3/metabolismo , Adulto , Cognición/fisiología , Cuerpo Estriado/diagnóstico por imagen , Femenino , Humanos , Masculino , Tomografía de Emisión de Positrones , RaclopridaRESUMEN
Obtaining accurate distance or depth information in endoscopy is crucial for the effective utilization of navigation systems. However, due to space constraints, incorporating depth cameras into endoscopic systems is often impractical. Our goal is to estimate depth images directly from endoscopic images using deep learning. This study presents a three-step methodology for training a depth-estimation network model. Initially, simulated endoscopy images and corresponding depth maps are generated using Unity based on a colon surface model obtained from segmented computed tomography colonography data. Subsequently, a cycle generative adversarial network model is employed to enhance the realism of the simulated endoscopy images. Finally, a deep learning model is trained using the synthesized endoscopy images and depth maps to estimate depths accurately. The performance of the proposed approach is evaluated and compared against prior studies utilizing unsupervised training methods. The results demonstrate the superior precision of the proposed technique in estimating depth images within endoscopy. The proposed depth estimation method holds promise for advancing the field by enabling enhanced navigation, improved lesion marking capabilities, and ultimately leading to better clinical outcomes.
RESUMEN
In many neuroscience and clinical studies, accurate measurement of hippocampus is very important to reveal the inter-subject anatomical differences or the subtle intra-subject longitudinal changes due to aging or dementia. Although many automatic segmentation methods have been developed, their performances are still challenged by the poor image contrast of hippocampus in the MR images acquired especially from 1.5 or 3.0 Tesla (T) scanners. With the recent advance of imaging technology, 7.0 T scanner provides much higher image contrast and resolution for hippocampus study. However, the previous methods developed for segmentation of hippocampus from 1.5 T or 3.0 T images do not work for the 7.0 T images, due to different levels of imaging contrast and texture information. In this paper, we present a learning-based algorithm for automatic segmentation of hippocampi from 7.0 T images, by taking advantages of the state-of-the-art multi-atlas framework and also the auto-context model (ACM). Specifically, ACM is performed in each atlas domain to iteratively construct sequences of location-adaptive classifiers by integrating both image appearance and local context features. Due to the plenty texture information in 7.0 T images, more advanced texture features are also extracted and incorporated into the ACM during the training stage. Then, under the multi-atlas segmentation framework, multiple sequences of ACM-based classifiers are trained for all atlases to incorporate the anatomical variability. In the application stage, for a new image, its hippocampus segmentation can be achieved by fusing the labeling results from all atlases, each of which is obtained by applying the atlas-specific ACM-based classifiers. Experimental results on twenty 7.0 T images with the voxel size of 0.35×0.35×0.35 mm3 show very promising hippocampus segmentations (in terms of Dice overlap ratio 89.1±0.020), indicating high applicability for the future clinical and neuroscience studies.
Asunto(s)
Algoritmos , Hipocampo/anatomía & histología , Interpretación de Imagen Asistida por Computador/métodos , Imagenología Tridimensional/métodos , Imagen por Resonancia Magnética/métodos , Modelos Anatómicos , Reconocimiento de Normas Patrones Automatizadas/métodos , Adulto , Inteligencia Artificial , Simulación por Computador , Femenino , Humanos , Aumento de la Imagen/métodos , Masculino , Modelos Neurológicos , Reproducibilidad de los Resultados , Sensibilidad y Especificidad , Técnica de SustracciónRESUMEN
The thalamus is one of the most important brain structures, with strong connections between subcortical and cortical areas of the brain. Most of the incoming information to the cortex passes through the thalamus. Accurate identification of substructures of the thalamus is therefore of great importance for the understanding of human brain connectivity. Direct visualization of thalamic substructures, however, is not easily achieved with currently available magnetic resonance imaging (MRI), including ultra-high field MRI such as 7.0T, mainly due to the limited contrast between the relevant structures. Recently, improvements in ultra-high field 7.0T MRI have opened the possibility of observing thalamic substructures by well-adjusted high-resolution T1 -weighted imaging. Moreover, the recently developed super-resolution track-density imaging (TDI) technique, based on results from whole-brain fiber-tracking, produces images with sub-millimeter resolution. These two methods enable us to show markedly improved anatomical detail of the substructures of the thalamus, including their detailed locations and directionality. In this study, we demonstrate the role of TDI for the visualization of the substructures of the thalamic nuclei, and relate these images to T1-weighted imaging at 7.0T MRI.
Asunto(s)
Imagen de Difusión Tensora/métodos , Imagen Eco-Planar/métodos , Imagenología Tridimensional/métodos , Núcleos Talámicos/ultraestructura , Adulto , Algoritmos , Femenino , Humanos , MasculinoRESUMEN
Advances in the new-generation of ultra-high-resolution, brain-dedicated positron emission tomography-magnetic resonance imaging (PET/MRI) systems have begun to provide many interesting insights into the molecular dynamics of the brain. First, the finely delineated structural information from ultra-high-field MRI can help us to identify accurate landmark structures, thereby making it easier to locate PET activation sites that are anatomically well-correlated with metabolic or ligand-specific organs in the neural structures in the brain. This synergistic potential of PET/MRI imaging is discussed in terms of neuroscience and neurological research from both translational and basic research perspectives. Experimental results from the hippocampus, thalamus, and brainstem obtained with (18)F-fluorodeoxyglucose and (11)C-3-amino-4-(2-dimethylaminomethylphenylsulfanyl)benzonitrile are used to demonstrate the potential of this new brain PET/MRI system.
Asunto(s)
Encefalopatías/diagnóstico , Imagen por Resonancia Magnética/métodos , Imagen Molecular/métodos , Tomografía de Emisión de Positrones/métodos , Compuestos de Anilina , Encefalopatías/diagnóstico por imagen , Radioisótopos de Carbono , Diseño de Equipo , Fluorodesoxiglucosa F18 , Humanos , Imagen por Resonancia Magnética/instrumentación , Imagen Molecular/instrumentación , Tomografía de Emisión de Positrones/instrumentación , Radiofármacos , SulfurosRESUMEN
We investigated the safety and clinical applicability of 7.0 Tesla (T) brain magnetic resonance imaging (MRI) in patients with brain tumors. Twenty-four patients with intraaxial or extraaxial brain tumors were enrolled in this study. 7.0T MRIs of T2*-weighted axial and T1-weighted coronal or sagittal images were obtained and compared with 1.5T brain MRIs. The T2*-weighted images from 7.0T brain MRI revealed detailed microvasculature and the internal contents of supratentorial brain tumors better than that of 1.5T brain MRI. For brain tumors located in parasellar areas or areas adjacent to major cerebral vessels, flow-related artifacts were exaggerated in the 7.0T brain MRIs. For brain tumors adjacent to the skull base, susceptibility artifacts in the interfacing areas of the paranasal sinus and skull base hampered the aquisition of detailed images and information on brain tumors in the 7.0T brain MRIs. This study shows that 7.0T brain MRI can provide detailed information on the intratumoral components and margins in supratentorial brain tumors. Further studies are needed to develop refined MRI protocols for better images of brain tumors located in the skull base, parasellar, and adjacent major cerebrovascular structures.
Asunto(s)
Neoplasias Encefálicas/diagnóstico por imagen , Adulto , Mareo/etiología , Femenino , Cefalea/etiología , Humanos , Imagen por Resonancia Magnética/efectos adversos , Masculino , Persona de Mediana Edad , Contracción Muscular/efectos de la radiación , RadiografíaRESUMEN
OBJECTIVES: This study was conducted to evaluate the accuracy of navigation process according to the type of tracking methods and registration markers. The target registration errors (TREs) were measured at seven anatomical landmarks of the mandible. METHODS: Four different experiments were performed to obtain the TREs using two tracking methods, the optical tracker (Polaris) and the electromagnetic (EM) tracker (Aurora), and two types of registration markers, invasive and noninvasive markers. All comparisons of TREs were statistically analyzed using SPSS and Python-based statistical package. RESULTS: The average TRE values obtained from the four experiments were as follows: (1) 0.85 mm (± 0.07) using invasive marker and Aurora, (2) 1.06 mm (± 0.12) using invasive marker and Polaris, (3) 1.43 mm (± 0.15) using noninvasive marker and Aurora, and (4) 1.57 mm (± 0.23) using noninvasive marker and Polaris. Comparisons between all the experimental results revealed statistically significant differences except for the type of tracking system. Although the comparison between the modality of the tracking system showed no significant differences, the EM-based approach consistently demonstrated better performances than the optical type in all comparisons. CONCLUSIONS: This study demonstrates that irrespective of the tracking modality, using invasive marker is a better choice in terms of accuracy. When using noninvasive marker, it is important to consider the increased TREs. In this study, the noninvasive marker caused a maximum increment of TREs of 0.81 mm compared with the invasive marker. Furthermore, using an EM-based tracker with invasive marker may result in the best accuracy for navigation.