Diagnostics

Research

Jump to: Review

20 pages, 1461 KiB

Open AccessArticle

Parallelistic Convolution Neural Network Approach for Brain Tumor Diagnosis

by Goodness Temofe Mgbejime, Md Altab Hossin, Grace Ugochi Nneji, Happy Nkanta Monday and Favour Ekong

Diagnostics 2022, 12(10), 2484; https://doi.org/10.3390/diagnostics12102484 - 13 Oct 2022

Cited by 6 | Viewed by 1950

Abstract

Today, Magnetic Resonance Imaging (MRI) is a prominent technique used in medicine, produces a significant and varied range of tissue contrasts in each imaging modalities, and is frequently employed by medical professionals to identify brain malignancies. With brain tumor being a very deadly [...] Read more.

Today, Magnetic Resonance Imaging (MRI) is a prominent technique used in medicine, produces a significant and varied range of tissue contrasts in each imaging modalities, and is frequently employed by medical professionals to identify brain malignancies. With brain tumor being a very deadly disease, early detection will help increase the likelihood that the patient will receive the appropriate medical care leading to either a full elimination of the tumor or the prolongation of the patient’s life. However, manually examining the enormous volume of magnetic resonance imaging (MRI) images and identifying a brain tumor or cancer is extremely time-consuming and requires the expertise of a trained medical expert or brain doctor to manually detect and diagnose brain cancer using multiple Magnetic Resonance images (MRI) with various modalities. Due to this underlying issue, there is a growing need for increased efforts to automate the detection and diagnosis process of brain tumor without human intervention. Another major concern most research articles do not consider is the low quality nature of MRI images which can be attributed to noise and artifacts. This article presents a Contrast Limited Adaptive Histogram Equalization (CLAHE) algorithm to precisely handle the problem of low quality MRI images by eliminating noisy elements and enhancing the visible trainable features of the image. The enhanced image is then fed to the proposed PCNN to learn the features and classify the tumor using sigmoid classifier. To properly train the model, a publicly available dataset is collected and utilized for this research. Additionally, different optimizers and different values of dropout and learning rates are used in the course of this study. The proposed PCNN with Contrast Limited Adaptive Histogram Equalization (CLAHE) algorithm achieved an accuracy of 98.7%, sensitivity of 99.7%, and specificity of 97.4%. In comparison with other state-of-the-art brain tumor methods and pre-trained deep transfer learning models, the proposed PCNN model obtained satisfactory performance. Full article

(This article belongs to the Special Issue Brain Imaging in Epilepsy)

► Show Figures

Figure 1

11 pages, 2845 KiB

Open AccessArticle

Electric Source Imaging in Presurgical Evaluation of Epilepsy: An Inter-Analyser Agreement Study

by Pietro Mattioli, Evy Cleeren, Levente Hadady, Alberto Cossu, Thomas Cloppenborg, Dario Arnaldi and Sándor Beniczky

Diagnostics 2022, 12(10), 2303; https://doi.org/10.3390/diagnostics12102303 - 24 Sep 2022

Cited by 2 | Viewed by 3363

Abstract

Electric source imaging (ESI) estimates the cortical generator of the electroencephalography (EEG) signals recorded with scalp electrodes. ESI has gained increasing interest for the presurgical evaluation of patients with drug-resistant focal epilepsy. In spite of a standardised analysis pipeline, several aspects tailored to [...] Read more.

Electric source imaging (ESI) estimates the cortical generator of the electroencephalography (EEG) signals recorded with scalp electrodes. ESI has gained increasing interest for the presurgical evaluation of patients with drug-resistant focal epilepsy. In spite of a standardised analysis pipeline, several aspects tailored to the individual patient involve subjective decisions of the expert performing the analysis, such as the selection of the analysed signals (interictal epileptiform discharges and seizures, identification of the onset epoch and time-point of the analysis). Our goal was to investigate the inter-analyser agreement of ESI in presurgical evaluations of epilepsy, using the same software and analysis pipeline. Six experts, of whom five had no previous experience in ESI, independently performed interictal and ictal ESI of 25 consecutive patients (17 temporal, 8 extratemporal) who underwent presurgical evaluation. The overall agreement among experts for the ESI methods was substantial (AC1 = 0.65; 95% CI: 0.59–0.71), and there was no significant difference between the methods. Our results suggest that using a standardised analysis pipeline, newly trained experts reach similar ESI solutions, calling for more standardisation in this emerging clinical application in neuroimaging. Full article

(This article belongs to the Special Issue Brain Imaging in Epilepsy)

► Show Figures

Graphical abstract

19 pages, 4544 KiB

Open AccessArticle

Novel User-Friendly Application for MRI Segmentation of Brain Resection following Epilepsy Surgery

by Roberto Billardello, Georgios Ntolkeras, Assia Chericoni, Joseph R. Madsen, Christos Papadelis, Phillip L. Pearl, Patricia Ellen Grant, Fabrizio Taffoni and Eleonora Tamilia

Diagnostics 2022, 12(4), 1017; https://doi.org/10.3390/diagnostics12041017 - 18 Apr 2022

Cited by 7 | Viewed by 3117

Abstract

Delineation of resected brain cavities on magnetic resonance images (MRIs) of epilepsy surgery patients is essential for neuroimaging/neurophysiology studies investigating biomarkers of the epileptogenic zone. The gold standard to delineate the resection on MRI remains manual slice-by-slice tracing by experts. Here, we proposed [...] Read more.

Delineation of resected brain cavities on magnetic resonance images (MRIs) of epilepsy surgery patients is essential for neuroimaging/neurophysiology studies investigating biomarkers of the epileptogenic zone. The gold standard to delineate the resection on MRI remains manual slice-by-slice tracing by experts. Here, we proposed and validated a semiautomated MRI segmentation pipeline, generating an accurate model of the resection and its anatomical labeling, and developed a graphical user interface (GUI) for user-friendly usage. We retrieved pre- and postoperative MRIs from 35 patients who had focal epilepsy surgery, implemented a region-growing algorithm to delineate the resection on postoperative MRIs and tested its performance while varying different tuning parameters. Similarity between our output and hand-drawn gold standards was evaluated via dice similarity coefficient (DSC; range: 0–1). Additionally, the best segmentation pipeline was trained to provide an automated anatomical report of the resection (based on presurgical brain atlas). We found that the best-performing set of parameters presented DSC of 0.83 (0.72–0.85), high robustness to seed-selection variability and anatomical accuracy of 90% to the clinical postoperative MRI report. We presented a novel user-friendly open-source GUI that implements a semiautomated segmentation pipeline specifically optimized to generate resection models and their anatomical reports from epilepsy surgery patients, while minimizing user interaction. Full article

(This article belongs to the Special Issue Brain Imaging in Epilepsy)

► Show Figures

Graphical abstract

15 pages, 1071 KiB

Open AccessArticle

Comparison of Qualitative and Quantitative Analyses of MR-Arterial Spin Labeling Perfusion Data for the Assessment of Pediatric Patients with Focal Epilepsies

by Domenico Tortora, Matteo Cataldi, Mariasavina Severino, Alessandro Consales, Mattia Pacetti, Costanza Parodi, Fiammetta Sertorio, Antonia Ramaglia, Erica Cognolato, Giulia Nobile, Margherita Mancardi, Giulia Prato, Laura Siri, Thea Giacomini, Pasquale Striano, Dario Arnaldi, Gianluca Piatelli, Andrea Rossi and Lino Nobili

Diagnostics 2022, 12(4), 811; https://doi.org/10.3390/diagnostics12040811 - 25 Mar 2022

Cited by 6 | Viewed by 3256

Abstract

The role of MR Arterial-Spin-Labeling Cerebral Blood Flow maps (ASL-CBF) in the assessment of pediatric focal epilepsy is still debated. We aim to compare the Seizure Onset Zone (SOZ) detection rate of three methods of evaluation of ASL-CBF: 1) qualitative visual (qCBF), 2) [...] Read more.

The role of MR Arterial-Spin-Labeling Cerebral Blood Flow maps (ASL-CBF) in the assessment of pediatric focal epilepsy is still debated. We aim to compare the Seizure Onset Zone (SOZ) detection rate of three methods of evaluation of ASL-CBF: 1) qualitative visual (qCBF), 2) z-score voxel-based quantitative analysis of index of asymmetry (AI-CBF), and 3) z-score voxel-based cluster analysis of the quantitative difference of patient’s CBF from the normative data of an age-matched healthy population (cCBF). Interictal ASL-CBF were acquired in 65 pediatric patients with focal epilepsy: 26 with focal brain lesions and 39 with a normal MRI. All hypoperfusion areas visible in at least 3 contiguous images of qCBF analysis were identified. In the quantitative evaluations, clusters with a significant z-score AI-CBF ≤ −1.64 and areas with a z-score cCBF ≤ −1.64 were considered potentially related to the SOZ. These areas were compared with the SOZ defined by the anatomo-electro-clinical data. In patients with a positive MRI, SOZ was correctly identified in 27% of patients using qCBF, 73% using AI-CBF, and 77% using cCBF. In negative MRI patients, SOZ was identified in 18% of patients using qCBF, in 46% using AI-CBF, and in 64% using cCBF (p < 0.001). Quantitative analyses of ASL-CBF maps increase the detection rate of SOZ compared to the qualitative method, principally in negative MRI patients. Full article

(This article belongs to the Special Issue Brain Imaging in Epilepsy)

► Show Figures

Figure 1

14 pages, 6436 KiB

Open AccessArticle

MRI in Late-Onset Rasmussen Encephalitis: A Long-Term Follow-Up Study

by Fabio Martino Doniselli, Francesco Deleo, Stefania Criscuolo, Andrea Stabile, Chiara Pastori, Roberta Di Giacomo, Giuseppe Didato, Luisa Chiapparini and Flavio Villani

Diagnostics 2022, 12(2), 502; https://doi.org/10.3390/diagnostics12020502 - 15 Feb 2022

Cited by 5 | Viewed by 3657

Abstract

Late-onset Rasmussen encephalitis (LoRE) is a rare unihemispheric progressive inflammatory disorder causing neurological deficits and epilepsy. The long-term radiological evolution has never been fully described. We retrospectively analyzed the MR images of 13 LoRE patients from a total of 136 studies, and searched [...] Read more.

Late-onset Rasmussen encephalitis (LoRE) is a rare unihemispheric progressive inflammatory disorder causing neurological deficits and epilepsy. The long-term radiological evolution has never been fully described. We retrospectively analyzed the MR images of 13 LoRE patients from a total of 136 studies, and searched for focal areas of volume loss or signal intensity abnormality in grey matter or white matter. Each subject had a median of nine MRI studies (IQR 7–13). Frontal and temporal lobes were the most affected regions (13/13 and 8/13, respectively) and showed the greatest worsening over time in terms of atrophic changes (9/13 and 5/8, respectively). A milder cortical atrophy was found in the insular and parietal lobes. The caudate nucleus was affected in seven patients. Hyperintensities of grey matter and white matter on T2-WI and FLAIR images were observed in all patients, and transiently in eight patients. In two cases out of the latter patients, these transient alterations evolved into atrophy of the same region. Disease duration was significantly associated with signal abnormalities in the grey matter at last follow-up. LoRE MRI alterations are milder, and their progression is markedly slower compared to radiological findings described in the childhood form. Full article

(This article belongs to the Special Issue Brain Imaging in Epilepsy)

► Show Figures

Figure 1

17 pages, 3532 KiB

Open AccessArticle

Comparison of In Vivo and Ex Vivo Magnetic Resonance Imaging in a Rat Model for Glioblastoma-Associated Epilepsy

by Charlotte Bouckaert, Emma Christiaen, Jeroen Verhoeven, Benedicte Descamps, Valerie De Meulenaere, Paul Boon, Evelien Carrette, Kristl Vonck, Christian Vanhove and Robrecht Raedt

Diagnostics 2021, 11(8), 1311; https://doi.org/10.3390/diagnostics11081311 - 21 Jul 2021

Cited by 2 | Viewed by 2656

Abstract

Magnetic resonance imaging (MRI) is frequently used for preclinical treatment monitoring in glioblastoma (GB). Discriminating between tumors and tumor-associated changes is challenging on in vivo MRI. In this study, we compared in vivo MRI scans with ex vivo MRI and histology to estimate [...] Read more.

Magnetic resonance imaging (MRI) is frequently used for preclinical treatment monitoring in glioblastoma (GB). Discriminating between tumors and tumor-associated changes is challenging on in vivo MRI. In this study, we compared in vivo MRI scans with ex vivo MRI and histology to estimate more precisely the abnormal mass on in vivo MRI. Epileptic seizures are a common symptom in GB. Therefore, we used a recently developed GB-associated epilepsy model from our group with the aim of further characterizing the model and making it useful for dedicated epilepsy research. Ten days after GB inoculation in rat entorhinal cortices, in vivo MRI (T2w and mean diffusivity (MD)), ex vivo MRI (T2w) and histology were performed, and tumor volumes were determined on the different modalities. The estimated abnormal mass on ex vivo T2w images was significantly smaller compared to in vivo T2w images, but was more comparable to histological tumor volumes, and might be used to estimate end-stage tumor volumes. In vivo MD images displayed tumors as an outer rim of hyperintense signal with a core of hypointense signal, probably reflecting peritumoral edema and tumor mass, respectively, and might be used in the future to distinguish the tumor mass from peritumoral edema—associated with reactive astrocytes and activated microglia, as indicated by an increased expression of immunohistochemical markers—in preclinical models. In conclusion, this study shows that combining imaging techniques using different structural scales can improve our understanding of the pathophysiology in GB. Full article

(This article belongs to the Special Issue Brain Imaging in Epilepsy)

► Show Figures

Figure 1

19 pages, 6492 KiB

Open AccessArticle

Changes in the Functional Brain Network of Children Undergoing Repeated Epilepsy Surgery: An EEG Source Connectivity Study

by Giulia Iandolo, Nitish Chourasia, Georgios Ntolkeras, Joseph R. Madsen, Christos Papadelis, Ellen Grant, Phillip L. Pearl, Fabrizio Taffoni and Eleonora Tamilia

Diagnostics 2021, 11(7), 1234; https://doi.org/10.3390/diagnostics11071234 - 9 Jul 2021

Cited by 12 | Viewed by 4393

Abstract

About 30% of children with drug-resistant epilepsy (DRE) continue to have seizures after epilepsy surgery. Since epilepsy is increasingly conceptualized as a network disorder, understanding how brain regions interact may be critical for planning re-operation in these patients. We aimed to estimate functional [...] Read more.

About 30% of children with drug-resistant epilepsy (DRE) continue to have seizures after epilepsy surgery. Since epilepsy is increasingly conceptualized as a network disorder, understanding how brain regions interact may be critical for planning re-operation in these patients. We aimed to estimate functional brain connectivity using scalp EEG and its evolution over time in patients who had repeated surgery (RS-group, n = 9) and patients who had one successful surgery (seizure-free, SF-group, n = 12). We analyzed EEGs without epileptiform activity at varying time points (before and after each surgery). We estimated functional connectivity between cortical regions and their relative centrality within the network. We compared the pre- and post-surgical centrality of all the non-resected (untouched) regions (far or adjacent to resection) for each group (using the Wilcoxon signed rank test). In alpha, theta, and beta frequency bands, the post-surgical centrality of the untouched cortical regions increased in the SF group (p < 0.001) whereas they decreased (p < 0.05) or did not change (p > 0.05) in the RS group after failed surgeries; when re-operation was successful, the post-surgical centrality of far regions increased (p < 0.05). Our data suggest that removal of the epileptogenic focus in children with DRE leads to a gain in the network centrality of the untouched areas. In contrast, unaltered or decreased connectivity is seen when seizures persist after surgery. Full article

(This article belongs to the Special Issue Brain Imaging in Epilepsy)

► Show Figures

Graphical abstract

Review

Jump to: Research

14 pages, 919 KiB

Open AccessReview

Brain Imaging in Epilepsy-Focus on Diffusion-Weighted Imaging

by Tzu-Hsin Huang, Ming-Chi Lai, Yu-Shiue Chen and Chin-Wei Huang

Diagnostics 2022, 12(11), 2602; https://doi.org/10.3390/diagnostics12112602 - 27 Oct 2022

Cited by 7 | Viewed by 2963

Abstract

Epilepsy is a common neurological disorder; 1% of people worldwide have epilepsy. Differentiating epileptic seizures from other acute neurological disorders in a clinical setting can be challenging. Approximately one-third of patients have drug-resistant epilepsy that is not well controlled by current antiepileptic drug [...] Read more.

Epilepsy is a common neurological disorder; 1% of people worldwide have epilepsy. Differentiating epileptic seizures from other acute neurological disorders in a clinical setting can be challenging. Approximately one-third of patients have drug-resistant epilepsy that is not well controlled by current antiepileptic drug therapy. Surgical treatment is potentially curative if the epileptogenic focus is accurately localized. Diffusion-weighted imaging (DWI) is an advanced magnetic resonance imaging technique that is sensitive to the diffusion of water molecules and provides additional information on the microstructure of tissue. Qualitative and quantitative analysis of peri-ictal, postictal, and interictal diffusion images can aid the differential diagnosis of seizures and seizure foci localization. This review focused on the fundamentals of DWI and its associated techniques, such as apparent diffusion coefficient, diffusion tensor imaging, and tractography, as well as their impact on epilepsy in terms of differential diagnosis, epileptic foci determination, and prognosis prediction. Full article

(This article belongs to the Special Issue Brain Imaging in Epilepsy)

► Show Figures

Figure 1

Journal Menu

Journal Browser

Brain Imaging in Epilepsy

Share This Special Issue

Special Issue Editor

Special Issue Information

Keywords

Benefits of Publishing in a Special Issue

Published Papers (8 papers)

Research

Review

Further Information

Guidelines

MDPI Initiatives

Follow MDPI