Tongue carcinoma: tumor volume measurement

Automatic Segmentation of MRI of Brain Tumor Using Deep Convolutional Network

Computer-aided diagnosis and treatment of multimodal magnetic resonance imaging (MRI) brain tumor image segmentation has always been a hot and significant topic in the field of medical image processing. Multimodal MRI brain tumor image segmentation utilizes the characteristics of each modal in the MRI image to segment the entire tumor and tumor core area and enhanced them from normal brain tissues. However, the grayscale similarity between brain tissues in various MRI images is very immense making it difficult to deal with the segmentation of multimodal MRI brain tumor images through traditional algorithms. Therefore, we employ the deep learning method as a tool to make full use of the complementary feature information between the multimodalities and instigate the following research: (i) build a network model suitable for brain tumor segmentation tasks based on the fully convolutional neural network framework and (ii) adopting an end-to-end training method, using two-dimensional slices of MRI images as network input data. The problem of unbalanced categories in various brain tumor image data is overcome by introducing the Dice loss function into the network to calculate the network training loss; at the same time, parallel Dice loss is proposed to further improve the substructure segmentation effect. We proposed a cascaded network model based on a fully convolutional neural network to improve the tumor core area and enhance the segmentation accuracy of the tumor area and achieve good prediction results for the substructure segmentation on the BraTS 2017 data set.

A Pilot Study: N-Staging Assessment of Shear Wave Elastrography in Small Cervical Lymph Nodes for Nasopharyngeal Carcinoma

Purpose: To investigate N-staging Assessment of pretreatment Shear wave elastrography (SWE) in small cervical lymph nodes (0. 5 cm ≤ maximum diameter < 1 cm, intact capsule, no central necrosis, sCLNs) in nasopharyngeal carcinoma (NPC) patients. Methods: Pathological biopsy proven 28 NPC patients with sCLNs shown in pretreatment magnetic resonance (MR) images and 40 target lymph nodes were enrolled. All target lymph nodes were divided into metastasis and benign lymph node groups according to pathology. SWE was used to exam the real time SWE imaging of each target lymph nodes before conducting ultrasonography guided fine needle biopsy. The minimum (Emin), maximum (Emax), and mean (Emean) elasticity indices (kPa) of target lymph nodes were recorded. The SWE examination was repeated three times for the same target lymph node and each elasticity indices for statistic was determined by average of three measurements. SPSS 21.0 statistics software is used for statistical analysis. The receiver operating characteristic (ROC) curve was performed to obtain the cutoff value of elasticity indices of metastatic sCLNs. Statistical significance was assumed when the P < 0.05. Results: Nine lymph nodes were metastatic and 31 were benign. The Emin, Emax, and Emean of benign group were 8.15 ± 6.12, 25.05 ± 12.37, and 16.05 ± 8.29 kPa, respectively; Emin, Emax, and Emean of metastasis group were 11.5 ± 6.17, 41.38 ± 17.87, and 23.48 ± 6.50 kPa, respectively. The difference of the Emax and Emean between metastasis and benign group were statistically significant (P = 0.003 and 0.018). The area under the ROC curve of Emin, Emax, and Emean of metastasis lymph node were 0.685 (P = 0.095), 0.785 (P = 0.010), and 0.765 (P = 0.017), respectively. Emax of 27 kPa and Emean of 17 kPa were taken as the cutoff value of diagnosis for metastasis sCLNs: the sensitivity, specificity, and accuracy were 77.8 and 100%, 71.0 and 61.3%, 75.0 and 70.0%, respectively. Conclusions: Pretreatment SWE has high accuracy in evaluating the sCLNs in NPC patients and is helpful for accurate N-staging and survival prognosis. It can be used as a clinical supplementary examination.

Validation of a Magnetic Resonance Imaging-based Auto-contouring Software Tool for Gross Tumour Delineation in Head and Neck Cancer Radiotherapy Planning

AIMS

To carry out statistical validation of a newly developed magnetic resonance imaging (MRI) auto-contouring software tool for gross tumour volume (GTV) delineation in head and neck tumours to assist in radiotherapy planning.

MATERIALS AND METHODS

Axial MRI baseline scans were obtained for 10 oropharyngeal and laryngeal cancer patients. GTV was present on 102 axial slices and auto-contoured using the modified fuzzy c-means clustering integrated with the level set method (FCLSM). Peer-reviewed (C-gold) manual contours were used as the reference standard to validate auto-contoured GTVs (C-auto) and mean manual contours (C-manual) from two expert clinicians (C1 and C2). Multiple geometric metrics, including the Dice similarity coefficient (DSC), were used for quantitative validation. A DSC≥0.7 was deemed acceptable. Inter- and intra-variabilities among the manual contours were also validated. The two-dimensional contours were then reconstructed in three dimensions for GTV volume calculation, comparison and three-dimensional visualisation.

RESULTS

The mean DSC between C-gold and C-auto was 0.79. The mean DSC between C-gold and C-manual was 0.79 and that between C1 and C2 was 0.80. The average time for GTV auto-contouring per patient was 8 min (range 6-13 min; mean 45 s per axial slice) compared with 15 min (range 6-23 min; mean 88 s per axial slice) for C1. The average volume concordance between C-gold and C-auto volumes was 86.51% compared with 74.16% between C-gold and C-manual. The average volume concordance between C1 and C2 volumes was 86.82%.

CONCLUSIONS

This newly designed MRI-based auto-contouring software tool shows initial acceptable results in GTV delineation of oropharyngeal and laryngeal tumours using FCLSM. This auto-contouring software tool may help reduce inter- and intra-variability and can assist clinical oncologists with time-consuming, complex radiotherapy planning.

Alpha shape theory for 3D visualization and volumetric measurement of brain tumor progression using magnetic resonance images

Resection of brain tumors is a tricky task in surgery due to its direct influence on the patients' survival rate. Determining the tumor resection extent for its complete information via-à-vis volume and dimensions in pre- and post-operative Magnetic Resonance Images (MRI) requires accurate estimation and comparison. The active contour segmentation technique is used to segment brain tumors on pre-operative MR images using self-developed software. Tumor volume is acquired from its contours via alpha shape theory. The graphical user interface is developed for rendering, visualizing and estimating the volume of a brain tumor. Internet Brain Segmentation Repository dataset (IBSR) is employed to analyze and determine the repeatability and reproducibility of tumor volume. Accuracy of the method is validated by comparing the estimated volume using the proposed method with that of gold-standard. Segmentation by active contour technique is found to be capable of detecting the brain tumor boundaries. Furthermore, the volume description and visualization enable an interactive examination of tumor tissue and its surrounding. Admirable features of our results demonstrate that alpha shape theory in comparison to other existing standard methods is superior for precise volumetric measurement of tumor.

Segmentation of tongue muscles from super-resolution magnetic resonance images

Imaging and quantification of tongue anatomy is helpful in surgical planning, post-operative rehabilitation of tongue cancer patients, and studying of how humans adapt and learn new strategies for breathing, swallowing and speaking to compensate for changes in function caused by disease, medical interventions or aging. In vivo acquisition of high-resolution three-dimensional (3D) magnetic resonance (MR) images with clearly visible tongue muscles is currently not feasible because of breathing and involuntary swallowing motions that occur over lengthy imaging times. However, recent advances in image reconstruction now allow the generation of super-resolution 3D MR images from sets of orthogonal images, acquired at a high in-plane resolution and combined using super-resolution techniques. This paper presents, to the best of our knowledge, the first attempt towards automatic tongue muscle segmentation from MR images. We devised a database of ten super-resolution 3D MR images, in which the genioglossus and inferior longitudinalis tongue muscles were manually segmented and annotated with landmarks. We demonstrate the feasibility of segmenting the muscles of interest automatically by applying the landmark-based game-theoretic framework (GTF), where a landmark detector based on Haar-like features and an optimal assignment-based shape representation were integrated. The obtained segmentation results were validated against an independent manual segmentation performed by a second observer, as well as against B-splines and demons atlasing approaches. The segmentation performance resulted in mean Dice coefficients of 85.3%, 81.8%, 78.8% and 75.8% for the second observer, GTF, B-splines atlasing and demons atlasing, respectively. The obtained level of segmentation accuracy indicates that computerized tongue muscle segmentation may be used in surgical planning and treatment outcome analysis of tongue cancer patients, and in studies of normal subjects and subjects with speech and swallowing problems.

State of the art survey on MRI brain tumor segmentation

Brain tumor segmentation consists of separating the different tumor tissues (solid or active tumor, edema, and necrosis) from normal brain tissues: gray matter (GM), white matter (WM), and cerebrospinal fluid (CSF). In brain tumor studies, the existence of abnormal tissues may be easily detectable most of the time. However, accurate and reproducible segmentation and characterization of abnormalities are not straightforward. In the past, many researchers in the field of medical imaging and soft computing have made significant survey in the field of brain tumor segmentation. Both semiautomatic and fully automatic methods have been proposed. Clinical acceptance of segmentation techniques has depended on the simplicity of the segmentation, and the degree of user supervision. Interactive or semiautomatic methods are likely to remain dominant in practice for some time, especially in these applications where erroneous interpretations are unacceptable. This article presents an overview of the most relevant brain tumor segmentation methods, conducted after the acquisition of the image. Given the advantages of magnetic resonance imaging over other diagnostic imaging, this survey is focused on MRI brain tumor segmentation. Semiautomatic and fully automatic techniques are emphasized.

Evaluating lingual carcinoma for surgical management: what does volumetric measurement with MRI offer?

MRI plays a crucial but under utilized role in the surgical management of lingual squamous cell carcinoma (SCC). Measurement of three-dimensional tumour volume (TV) has the potential to guide management of clinically negative cervical lymph nodes and address deficiencies in current TNM staging criteria This work studied the value of MRI-measured TV as a predictor of 2 year disease-related survival (DRS) and disease-free survival (DFS), as well as occult cervical lymph node metastasis (OM) in lingual cancer. TV was determined by manually segmenting the tumour contour in each image slice and using the resulting pixel value to calculate the three-dimensional extent of disease. TV was also compared with the more established measure of tumour thickness (TT) Significant differences in DRS (χ²(1) = 7.7, Hazard ratio (HR) = 7.3, p = 0.005) and DFS (χ²(1) = 5.6, HR = 4.3, p = 0.02) at two years were found using a cut-off of 8 cm³. Similarly, a significant relationship between TV and occult cervical lymph node metastasis was discovered using a 3 cm³ cut-off (OR = 6.7, p = 0.02, Fisher's Exact Test).

Temporal Changes in IMRT Contouring of Organs at Risk for Nasopharyngeal Carcinoma — The Learning Curve Blues and a Tool that Could Help

With improved target conformality, the transition to IMRT for nasopharyngeal cancer (NPC) has moved contouring accuracy and consistency to the forefront. At NUH, IMRT for NPC was implemented in 2005, with more than 70 patients treated since then. The objective was to measure the accuracy and variability of contouring organs at risk (OAR) over time. The first 10 patients, 5 from each of the two head and neck (H&N) Radiation Oncologists, treated by IMRT in 2005 formed cohort A. Ten patients, matched by stage, treated by IMRT in 2007 formed cohort B. The RTOG 0225 protocol was followed. These 20 plans were retrieved from archive. A H&N Radiologist, who is a member of the UICC Expert Panel for the TNM Staging of NPC, reviewed the original OAR contours and developed a standardized OAR contouring template. Using the template, the OAR volumes were then re-contoured in all 20 cases, representing the gold standard against which the original volume was compared. For each patient, comparisons were made between the original and standardized contours using volumetric and spatial parameters. Cohort A was then compared with cohort B to determine whether accuracy and variability changed over time. Evaluated OAR volumes included the temporal lobes, brainstem, optic nerves, optic chiasm, pituitary, temporo-mandibular joint (TMJ), parotid glands, inner ears, eyes, and thyroid. While the original temporal lobe contours were significantly larger in cohort B (60.2 vs. 106.8 mL, p=0.02), the absolute difference between the original and standardized volumes was reduced by 53% (p=0.02) and there was no difference in the centroid coordinates and the overlapping fraction. While the inner ear was consistently contoured between cohort A and B, there was systematic exclusion of the cochlea in the contours. The original optic nerve contours decreased from cohort A to B (p=0.008), with an improvement in overlap fraction (p=0.06). The TMJ original volumes were smaller for cohort B than A, with a correspondingly significant improvement in overlapping fraction (p=0.02) with the standardized volumes. No difference was seen in the remaining OAR.

SIGNIFICANCE OF TUMOUR VOLUME MEASUREMENTS IN TONGUE CANCER: A NOVEL ROLE IN STAGING

BACKGROUND

Tongue cancers are staged by the American Joint Committee on Cancer and the Union Internationale Contre le Cancer TNM staging systems. Cancer, however, evolves in a 3-D plane. Hence, using the largest tumour diameter will not reflect total cancer volume. We aim to evaluate the use of tongue cancer tumour volume (Tv) as a prognostic predictor of disease recurrence and survival.

METHODS

The study is a retrospective analysis of patients in Singapore General Hospital who underwent complete resection for histologically proven tongue carcinoma from 2000 to 2002. The Tv was measured on staging T(2)-weighted magnetic resonance imaging datasets by semiautomated methods.

RESULTS

Seventeen patients with a median follow-up duration of 57.9 months were studied. A wide range of volumes was noted in each T stage. The median time to relapse was 8.6 months for those with Tv > or = 13 cc but was not achieved for those with Tv < 13 cc. The hazard ratio comparing Tv > or = 13 cc versus <13 cc is 9.02 (95% confidence interval (CI) 1.70-47.94, P = 0.014). Of the seven deaths reported, five patients had Tv > or = 13 cc. The median overall survival was 15.8 months for those with Tv > or = 13 cc but was not achieved for those with Tv < 13 cc. The hazards of death for Tv > or = 13 cc was 3.91 times that of Tv < 13 cc (95% CI 0.86-17.86, P = 0.078).

CONCLUSION

Tongue cancer Tv measurement allows a more refined and accurate assessment of tumour status. This can be a possible prognostic indicator and be used in a novel staging method for the future.

Imaging of tongue carcinoma

The tongue enables taste and plays a critical role in formation of food bolus and deglutition. The tongue is also crucial for speech and the earliest sign of tongue paresis is a change in the quality of speech. Given the importance of the tongue, tongue carcinoma should be accurately staged in order to optimise treatment options and preserve organ function. The intent of this review is to familiarise radiologists with the pertinent anatomy of the tongue and the behaviour of tongue carcinoma so as to map malignant infiltration accurately.

CT-Based Evaluation of Tumor Volume After Intra-Arterial Chemotherapy of Locally Advanced Carcinoma of the Oral Cavity: Comparison with Clinical Remission Rates

PURPOSE

To assess the volume of locally advanced tumors of the oral cavity and the oropharynx before and after intra-arterial (i.a.) chemotherapy by means of computed tomography and to compare these data with clinically determined treatment response of the same patient population.

METHODS

Eighty-eight patients with histologically proven, advanced carcinoma of the oral cavity and/or the oropharynx (local tumor stages T3/4) received neoadjuvant i.a. chemotherapy with cisplatin as part of a multimodal therapeutic regimen, comprising (1) local chemotherapy, (2) surgery, and (3) combined radio-chemotherapy. Three weeks after the intervention, residual disease was evaluated radiologically by measurement of the tumor volume and clinically by inspection and palpation of the primary tumor according to WHO criteria.

RESULTS

Comparison of treatment response according to radiological and clinical criteria respectively revealed complete remission in 5% vs. 8% (p < 0.05), partial remission in 30% vs. 31%, stable disease in 61% vs. 58%, and tumor progression in 5% vs. 2%.

CONCLUSION

Radiological volumetry and clinical evaluation found comparable response rates after local chemotherapy. However, in patients with good response after local treatment, volumetric measurement with CT may help to distinguish between partial and complete remission. Thus, radiological tumor volumetry provides precise and differentiated information about tumor response and should be used as an additional tool in treatment monitoring after local chemotherapy.

Reliability of CT-based tumor volumetry after intraarterial chemotherapy in patients with small carcinoma of the oral cavity and the oropharynx

INTRODUCTION

The aim of the study was to evaluate the feasibility and consistency of CT-based tumor volumetry in patients with early carcinoma of the oral cavity and the oropharynx before and after intraarterial (IA) chemotherapy, comparing these data with clinical remission rates.

METHODS

Included in the study were 61 patients (mean age 59.3 years; 47 men) with histologically proven small carcinoma of the oral cavity or the oropharynx (local tumor stages T1/2). Patients received IA chemotherapy with high-dose cisplatin as part of a multimodal therapeutic regimen and underwent both clinical and radiological examination before and 4 weeks after local chemotherapy.

RESULTS

Clinical evaluation of tumor response was possible in all patients (61/61). Radiological assessment of tumor volume was feasible in 42 of 61 patients (69%), but failed in 19 (31%) due to the absence of deep tumoral spread, lack of contrast enhancement or severe dental artifacts. Patients in whom evaluation was possible according to volumetric and clinical criteria revealed comparable remission rates: overall response 54.8% versus 52.4%, stable disease 40.4% versus 47.6%, and tumor progression 4.8% versus 0.0%.

CONCLUSION

Because volume calculation was not feasible in approximately one-third of the patients, it cannot be recommended as a reliable indicator for treatment response in patients with small carcinoma of the oral cavity.

Correlation between MR imaging–derived nasopharyngeal carcinoma tumor volume and TNM system

PURPOSE

To measure nasopharyngeal carcinoma tumor volume based on magnetic resonance images using a validated semiautomated measurement methodology and correlate tumor volume with TNM T classification.

METHODS AND MATERIALS

The study population consisted of 206 consecutive nasopharyngeal carcinoma patients who had magnetic resonance imaging staging scans. Tumor volume was measured using a semisupervised knowledge-based fuzzy clustering algorithm. Patients were divided into 4 groups according to TNM T classification. The difference in tumor volumes among the various TNM T-classification groups was examined.

RESULTS

The mean tumor volume in each T-classification group is as follows: T1, 8.6 mL +/- 5.0 (standard deviation [SD]); T2, 18.1 mL +/- 8.1 (SD); T3, 25.8 mL +/- 14.1 (SD); and T4, 36.2 mL +/- 18.9 (SD). The mean tumor volume increased significantly with advancing T classification (p < 0.0001). Tumor volume in a more advanced T group was significantly larger than that in an adjacent early T group (p < 0.01).

CONCLUSION

Validated magnetic resonance imaging-based tumor volume shows positive correlation between tumor volume and advancing T-classification groups. It may be possible to incorporate tumor volume as an additional prognostic factor into the existing TNM system.

Enlargement of post-traumatic intracerebral haematoma: incidence and time course

We retrospectively assessed the incidence and time course of enlargement in post-traumatic intracerebral haematoma (PTICH). Computed tomography (CT) scans from 165 patients who underwent a scan within 72 h and a repeat scan within 120 h of the onset of trauma were examined. A semi-automated method using region deformation-based segmentation was used to calculate the haematoma volume. The presence of haematoma enlargement was also determined based on a consensus by five observers. Seventy cases (42%) showed enlargement of the haematoma. The frequency of haematoma enlargement decreased as the interval between the onset of trauma and the initial scan increased. The discriminant value of the ratio of the haematoma volume in the second scan to that in the initial scan was ascertained, and the cut-off value for haematoma enlargement was determined to be 1.45. The radiographic criterion for enlargement in PTICH on CT scan was, therefore, defined as a > or = 1.45 times increase in haematoma volume.