Lingfeng Wen

The topography and significance of extratemporal hypometabolism in refractory mesial temporal lobe epilepsy examined by FDG‐PET

Purpose:  This study aims to map the temporal and extratemporal 18‐fluorodeoxyglucose positron emission tomography (FDG‐PET)–defined hypometabolism in mesial temporal lobe epilepsy (MTLE). We hypothesize that quantitative analysis will reveal extensive extratemporal glucose hypometabolism (EH), that the EH is related to seizure propagation beyond the temporal lobe, hypometabolism restricted to one temporal lobe predicts a good outcome following surgery, and EH predicts a poor outcome.

3D neurological image retrieval with localized pathology-centric CMRGlc patterns

Functional neuroimaging has an important role in non-invasive diagnosis of neurodegenerative disorders. There are now large volumes of imaging data generated by functional imaging technologies and so there is a need to efficiently manage and retrieve these data. In this paper, we propose a new scheme for efficient 3D content-based neurological image retrieval. 3D pathology-centric masks were adaptively designed and applied for extracting CMRGlc (cerebral metabolic rate of glucose consumption) texture features with volumetric co-occurrence matrices from neurological FDG PET images. Our results, using 93 clinical dementia studies, show that our approach offers a robust and efficient retrieval mechanism for relevant clinical cases and provides advantages in image data analysis and management.

Localized functional neuroimaging retrieval using 3D discrete curvelet transform

Neuroimaging is a fundamental component of the neurological diagnosis. The greatly increased volume and complexity of neuroimaging datasets has created a need for efficient image management and retrieval. In this paper, we advance a content-based retrieval framework for 3D functional neuroimaging data based on 3D curvelet transforms. The localized volumetric texture feature was extracted by a 3D digital curvelet transform from parametric image of cerebral metabolic rate of glucose consumption with a set of adaptive disorder-oriented masks for each type of neurological disorder. The results, using 142 clinical dementia studies, show that our proposed approach supports efficient and high performance neuroimaging data retrieval.

Multi-Channel neurodegenerative pattern analysis and its application in Alzheimer's disease characterization.

Neuroimaging has played an important role in non-invasive diagnosis and differentiation of neurodegenerative disorders, such as Alzheimers disease and Mild Cognitive Impairment. Various features have been extracted from the neuroimaging data to characterize the disorders, and these features can be roughly divided into global and local features. Recent studies show a tendency of using local features in disease characterization, since they are capable of identifying the subtle disease-specific patterns associated with the effects of the disease on human brain. However, problems arise if the neuroimaging database involved multiple disorders or progressive disorders, as disorders of different types or at different progressive stages might exhibit different degenerative patterns. It is difficult for the researchers to reach consensus on what brain regions could effectively distinguish multiple disorders or multiple progression stages. In this study we proposed a Multi-Channel pattern analysis approach to identify the most discriminative local brain metabolism features for neurodegenerative disorder characterization. We compared our method to global methods and other pattern analysis methods based on clinical expertise or statistics tests. The preliminary results suggested that the proposed Multi-Channel pattern analysis method outperformed other approaches in Alzheimers disease characterization, and meanwhile provided important insights into the underlying pathology of Alzheimers disease and Mild Cognitive Impairment.

A graph-based approach for the retrieval of multi-modality medical images

In this paper, we address the retrieval of multi-modality medical volumes, which consist of two different imaging modalities, acquired sequentially, from the same scanner. One such example, positron emission tomography and computed tomography (PET-CT), provides physicians with complementary functional and anatomical features as well as spatial relationships and has led to improved cancer diagnosis, localisation, and staging. The challenge of multi-modality volume retrieval for cancer patients lies in representing the complementary geometric and topologic attributes between tumours and organs. These attributes and relationships, which are used for tumour staging and classification, can be formulated as a graph. It has been demonstrated that graph-based methods have high accuracy for retrieval by spatial similarity. However, naïvely representing all relationships on a complete graph obscures the structure of the tumour-anatomy relationships. We propose a new graph structure derived from complete graphs that structurally constrains the edges connected to tumour vertices based upon the spatial proximity of tumours and organs. This enables retrieval on the basis of tumour localisation. We also present a similarity matching algorithm that accounts for different feature sets for graph elements from different imaging modalities. Our method emphasises the relationships between a tumour and related organs, while still modelling patient-specific anatomical variations. Constraining tumours to related anatomical structures improves the discrimination potential of graphs, making it easier to retrieve similar images based on tumour location. We evaluated our retrieval methodology on a dataset of clinical PET-CT volumes. Our results showed that our method enabled the retrieval of multi-modality images using spatial features. Our graph-based retrieval algorithm achieved a higher precision than several other retrieval techniques: gray-level histograms as well as state-of-the-art methods such as visual words using the scale- invariant feature transform (SIFT) and relational matrices representing the spatial arrangements of objects.

A supervised multiview spectral embedding method for neuroimaging classification

The multi-view/multi-modal features are commonly used in neuroimaging classification because they could provide complementary information to each other and thus result in better classification performance than single-view features. However, it is very challenging to effectively integrate such rich features, since straightforward concatenation or singleview spectral embedding methods rarely leads to physically meaningful integration. In this paper, we present a supervised multi-view/multi-modal spectral embedding method (SMSE) for neuroimaging classification. This method embeds the high dimensional multi-view features derived from multi-modal neuroimaging data into a low dimensional feature space and preserves the optimal local embeddings among different views. The proposed SMSE algorithm, validated using three groups of neuroimaging data, is able to achieve significant classification improvement over the state-of-the-art multi-view spectral embedding methods.

Multi-Channel brain atrophy pattern analysis in neuroimaging retrieval

The high throughput 3D neuroimaging datasets have posed great challenges for neuroimaging data retrieval. To achieve more accurate neuroimaging retrieval, various content-based retrieval approaches have been proposed. Recent studies showed a tendency of using the localized features extracted from a subset of the brain structures, instead of the global features extracted from whole brain. However, these studies relied heavily on specific pattern analysis techniques or clinical expertise. In this study we proposed a Multi-Channel pattern analysis approach to identify the most discriminative disease-sensitive brain structures for neurodegenerative disorders and thus to enhance neuroimaging retrieval. The preliminary results suggested that the proposed Multi-Channel pattern analysis approach could confidently identify the brain structures with atrophy and further improve the neuroimaging retrieval performance.

Cellular automata and anisotropic diffusion filter based interactive tumor segmentation for positron emission tomography

Tumor segmentation in positron emission tomography (PET) aids clinical diagnosis and in assessing treatment response. However, the low resolution and signal-to-noise inherent in PET images, makes accurate tumor segmentation challenging. Manual delineation is time-consuming and subjective, whereas fully automated algorithms are often limited to particular tumor types, and have difficulties in segmenting small and low-contrast tumors. Interactive segmentation may reduce the inter-observer variability and minimize the user input. In this study, we present a new interactive PET tumor segmentation method based on cellular automata (CA) and a nonlinear anisotropic diffusion filter (ADF). CA is tolerant of noise and image pattern complexity while ADF reduces noise while preserving edges. By coupling CA with ADF, our proposed approach was robust and accurate in detecting and segmenting noisy tumors. We evaluated our method with computer simulation and clinical data and it outperformed other common interactive PET segmentation algorithms.

Use of anatomical priors in the segmentation of PET lung tumor images

Advances in dual-modality imaging that combine anatomical and functional images have considerably improved tumor staging and treatment planning. PET/CT can, for example, detect tumor invasion into adjacent tissues, as well as provide precise localization of lesions, even when no morphological changes are identified by CT. In lung cancer staging and therapy planning, determination of a tumors size, its invasion into adjacent structures, mediastinal node status, and the detection of distant metastases are of great importance. In this study, we investigated the use of anatomical priors in the segmentation of tumors in simulated functional PET images of the lungs. The anatomical information was used as priors to extract the lung structures from the co-aligned PET data. The performance of a conventional iterative pixel-classification algorithm of fuzzy c-means (FCM) cluster analysis for segmenting the PET data with and without the use of the priors was quantitatively evaluated. A Monte Carlo simulation of PET with anatomical priors derived from the Zubal whole-body phantom was used in the evaluation. We demonstrate that the use of the anatomical priors to restrict the PET data to regions of interest consisting only of lung structures is able to improve the accuracy and reliability of the cluster analysis segmentation of lung tumors in PET images.

Generalized regional disorder-sensitive-weighting scheme for 3D neuroimaging retrieval

3D functional neuroimaging is used in the diagnosis and management of neurological disorders. The efficient management and analysis of these large imaging datasets has prompted research in the field of content-based image retrieval. In this context, our generalized regional disorder-sensitive-weighting (DSW) scheme gives greater weight to brain regions affected by the diseases than regions that are relatively spared. We used two DSW matrices; one matrix is based on the occurrence maps that highlight abnormal functional regions; the other is based on the regional Fisher discriminant ratio. Our results suggest that our DSW matrices enhance neuroimaging data retrieval and provide a flexible weighting solution for the clinical analysis of different types of neurological disorders.