MRI measurement of brain tumor response: comparison of visual metric and automatic segmentation

Direct Evaluation of Treatment Response in Brain Metastatic Disease with Deep Neuroevolution

Cancer centers have an urgent and unmet clinical and research need for AI that can guide patient management. A core component of advancing cancer treatment research is assessing response to therapy. Doing so by hand, for example, as per RECIST or RANO criteria, is tedious and time-consuming, and can miss important tumor response information. Most notably, the prevalent response criteria often exclude lesions, the non-target lesions, altogether. We wish to assess change in a holistic fashion that includes all lesions, obtaining simple, informative, and automated assessments of tumor progression or regression. Because genetic sub-types of cancer can be fairly specific and patient enrollment in therapy trials is often limited in number and accrual rate, we wish to make response assessments with small training sets. Deep neuroevolution (DNE) is a novel radiology artificial intelligence (AI) optimization approach that performs well on small training sets. Here, we use a DNE parameter search to optimize a convolutional neural network (CNN) that predicts progression versus regression of metastatic brain disease. We analyzed 50 pairs of MRI contrast-enhanced images as our training set. Half of these pairs, separated in time, qualified as disease progression, while the other 25 image pairs constituted regression. We trained the parameters of a CNN via "mutations" that consisted of random CNN weight adjustments and evaluated mutation "fitness" as summed training set accuracy. We then incorporated the best mutations into the next generation's CNN, repeating this process for approximately 50,000 generations. We applied the CNNs to our training set, as well as a separate testing set with the same class balance of 25 progression and 25 regression cases. DNE achieved monotonic convergence to 100% training set accuracy. DNE also converged monotonically to 100% testing set accuracy. We have thus shown that DNE can accurately classify brain metastatic disease progression versus regression. Future work will extend the input from 2D image slices to full 3D volumes, and include the category of "no change." We believe that an approach such as ours can ultimately provide a useful and informative complement to RANO/RECIST assessment and volumetric AI analysis.

A Brief History of Machine Learning in Neurosurgery

The history of machine learning in neurosurgery spans three decades and continues to develop at a rapid pace. The earliest applications of machine learning within neurosurgery were first published in the 1990s as researchers began developing artificial neural networks to analyze structured datasets and supervised tasks. By the turn of the millennium, machine learning had evolved beyond proof-of-concept; algorithms had success detecting tumors in unstructured clinical imaging, and unsupervised learning showed promise for tumor segmentation. Throughout the 2000s, the role of machine learning in neurosurgery was further refined. Well-trained models began to consistently best expert clinicians at brain tumor diagnosis. Additionally, the digitization of the healthcare industry provided ample data for analysis, both structured and unstructured. By the 2010s, the use of machine learning within neurosurgery had exploded. The rapid deployment of an exciting new toolset also led to the growing realization that it may offer marginal benefit at best over conventional logistical regression models for analyzing tabular datasets. Additionally, the widespread adoption of machine learning in neurosurgical clinical practice continues to lag until additional validation can ensure generalizability. Many exciting contemporary applications nonetheless continue to demonstrate the unprecedented potential of machine learning to revolutionize neurosurgery when applied to appropriate clinical challenges.

Neurosurgery and artificial intelligence

Neurosurgeons receive extensive and lengthy training to equip themselves with various technical skills, and neurosurgery require a great deal of pre-, intra- and postoperative clinical data collection, decision making, care and recovery. The last decade has seen a significant increase in the importance of artificial intelligence (AI) in neurosurgery. AI can provide a great promise in neurosurgery by complementing neurosurgeons' skills to provide the best possible interventional and noninterventional care for patients by enhancing diagnostic and prognostic outcomes in clinical treatment and help neurosurgeons with decision making during surgical interventions to improve patient outcomes. Furthermore, AI is playing a pivotal role in the production, processing and storage of clinical and experimental data. AI usage in neurosurgery can also reduce the costs associated with surgical care and provide high-quality healthcare to a broader population. Additionally, AI and neurosurgery can build a symbiotic relationship where AI helps to push the boundaries of neurosurgery, and neurosurgery can help AI to develop better and more robust algorithms. This review explores the role of AI in interventional and noninterventional aspects of neurosurgery during pre-, intra- and postoperative care, such as diagnosis, clinical decision making, surgical operation, prognosis, data acquisition, and research within the neurosurgical arena.

Automatic brain lesion segmentation on standard magnetic resonance images: a scoping review

OBJECTIVES

Medical image analysis practices face challenges that can potentially be addressed with algorithm-based segmentation tools. In this study, we map the field of automatic MR brain lesion segmentation to understand the clinical applicability of prevalent methods and study designs, as well as challenges and limitations in the field.

DESIGN

Scoping review.

SETTING

Three databases (PubMed, IEEE Xplore and Scopus) were searched with tailored queries. Studies were included based on predefined criteria. Emerging themes during consecutive title, abstract, methods and whole-text screening were identified. The full-text analysis focused on materials, preprocessing, performance evaluation and comparison.

RESULTS

Out of 2990 unique articles identified through the search, 441 articles met the eligibility criteria, with an estimated growth rate of 10% per year. We present a general overview and trends in the field with regard to publication sources, segmentation principles used and types of lesions. Algorithms are predominantly evaluated by measuring the agreement of segmentation results with a trusted reference. Few articles describe measures of clinical validity.

CONCLUSIONS

The observed reporting practices leave room for improvement with a view to studying replication, method comparison and clinical applicability. To promote this improvement, we propose a list of recommendations for future studies in the field.

Automatic segmentation of brain tumors in magnetic resonance imaging

Objective

To develop a computational algorithm applied to magnetic resonance imaging for automatic segmentation of brain tumors.

Methods

A total of 130 magnetic resonance images were used in the T1c, T2 and FSPRG T1C sequences and in the axial, sagittal and coronal planes of patients with brain cancer. The algorithms employed contrast correction, histogram normalization and binarization techniques to disconnect adjacent structures from the brain and enhance the region of interest. Automatic segmentation was performed through detection by coordinates and arithmetic mean of the area. Morphological operators were used to eliminate undesirable elements and reconstruct the shape and texture of the tumor. The results were compared with manual segmentations by two radiologists to determine the efficacy of the algorithms implemented.

Results

The correlated correspondence between the segmentation obtained and the gold standard was 89.23%.

Conclusion

It is possible to locate and define the tumor region automatically with no the need for user interaction, based on two innovative methods to detect brain extreme sites and exclude non-tumor tissues on magnetic resonance images.

Natural and Artificial Intelligence in Neurosurgery: A Systematic Review

BACKGROUND

Machine learning (ML) is a domain of artificial intelligence that allows computer algorithms to learn from experience without being explicitly programmed.

OBJECTIVE

To summarize neurosurgical applications of ML where it has been compared to clinical expertise, here referred to as "natural intelligence."

METHODS

A systematic search was performed in the PubMed and Embase databases as of August 2016 to review all studies comparing the performance of various ML approaches with that of clinical experts in neurosurgical literature.

RESULTS

Twenty-three studies were identified that used ML algorithms for diagnosis, presurgical planning, or outcome prediction in neurosurgical patients. Compared to clinical experts, ML models demonstrated a median absolute improvement in accuracy and area under the receiver operating curve of 13% (interquartile range 4-21%) and 0.14 (interquartile range 0.07-0.21), respectively. In 29 (58%) of the 50 outcome measures for which a P-value was provided or calculated, ML models outperformed clinical experts (P < .05). In 18 of 50 (36%), no difference was seen between ML and expert performance (P > .05), while in 3 of 50 (6%) clinical experts outperformed ML models (P < .05). All 4 studies that compared clinicians assisted by ML models vs clinicians alone demonstrated a better performance in the first group.

CONCLUSION

We conclude that ML models have the potential to augment the decision-making capacity of clinicians in neurosurgical applications; however, significant hurdles remain associated with creating, validating, and deploying ML models in the clinical setting. Shifting from the preconceptions of a human-vs-machine to a human-and-machine paradigm could be essential to overcome these hurdles.

Monitoring optic chiasmatic-hypothalamic glioma volumetric changes by MRI in children under clinical surveillance or chemotherapy

PURPOSE

Optic pathway gliomas represent 5% of pediatric brain tumors and are typically low-grade lesions. Because of their unpredictable clinical course, adequate treatment approaches have been controversial, involving surveillance, surgery, chemotherapy, and radiotherapy. In this study, we use volumetric imaging to compare evolution of optic chiasmatic-hypothalamic gliomas (OCHG) treated with and without chemotherapy, analyzing tumor volume variation during the overall period.

METHODS

A total of 45 brain MRI were retrospectively analyzed for 14 patients with OCHG. Volumetric assessment of the lesions was performed by a neuroradiologist, using software DISPLAY. OCHG patients were allocated into two groups: group 1 (n = 8) who underwent chemotherapy and group 2 (n = 6) who did not receive chemotherapy. Outcome analysis was performed comparing tumor volume evolution of these two groups.

RESULTS

The results showed a reduction of 4.4% of the volume of the lesions for group 1 after the end of chemotherapy, with an increase of 5.3% in volume in the late follow-up examination. For group 2, we found a slight reduction (5%) of the overall volume of the lesions, both with no statistical significance (p > 0.05).

CONCLUSIONS

From the limited series analyzed in this study, no significant differences were observed in relation to the volume change of lesions treated or not treated with chemotherapy. Larger prospective clinical trials are needed to better evaluate the effect of chemotherapy and radiological response of OCHG.

An introduction and overview of machine learning in neurosurgical care

BACKGROUND

Machine learning (ML) is a branch of artificial intelligence that allows computers to learn from large complex datasets without being explicitly programmed. Although ML is already widely manifest in our daily lives in various forms, the considerable potential of ML has yet to find its way into mainstream medical research and day-to-day clinical care. The complex diagnostic and therapeutic modalities used in neurosurgery provide a vast amount of data that is ideally suited for ML models. This systematic review explores ML's potential to assist and improve neurosurgical care.

METHOD

A systematic literature search was performed in the PubMed and Embase databases to identify all potentially relevant studies up to January 1, 2017. All studies were included that evaluated ML models assisting neurosurgical treatment.

RESULTS

Of the 6,402 citations identified, 221 studies were selected after subsequent title/abstract and full-text screening. In these studies, ML was used to assist surgical treatment of patients with epilepsy, brain tumors, spinal lesions, neurovascular pathology, Parkinson's disease, traumatic brain injury, and hydrocephalus. Across multiple paradigms, ML was found to be a valuable tool for presurgical planning, intraoperative guidance, neurophysiological monitoring, and neurosurgical outcome prediction.

CONCLUSIONS

ML has started to find applications aimed at improving neurosurgical care by increasing the efficiency and precision of perioperative decision-making. A thorough validation of specific ML models is essential before implementation in clinical neurosurgical care. To bridge the gap between research and clinical care, practical and ethical issues should be considered parallel to the development of these techniques.

A data-oriented self-calibration and robust chemical-shift encoding by using clusterization (OSCAR): Theory, optimization and clinical validation in neuromuscular disorders

Multi-echo Chemical Shift-Encoded (CSE) methods for Fat-Water quantification are growing in clinical use due to their ability to estimate and correct some confounding effects. State of the art CSE water/fat separation approaches rely on a multi-peak fat spectrum with peak frequencies and relative amplitudes kept constant over the entire MRI dataset. However, the latter approximation introduces a systematic error in fat percentage quantification in patients where the differences in lipid chemical composition are significant (such as for neuromuscular disorders) because of the spatial dependence of the peak amplitudes. The present work aims to overcome this limitation by taking advantage of an unsupervised clusterization-based approach offering a reliable criterion to carry out a data-driven segmentation of the input MRI dataset into multiple regions. Results established that the presented algorithm is able to identify at least 4 different partitions from MRI dataset under which to perform independent self-calibration routines and was found robust in NMD imaging studies (as evaluated on a cohort of 24 subjects) against latest CSE techniques with either calibrated or non-calibrated approaches. Particularly, the PDFF of the thigh was more reproducible for the quantitative estimation of pathological muscular fat infiltrations, which may be promising to evaluate disease progression in clinical practice.

Comparison of visual metric and planimetry methods for brain tumor measurement in dogs

OBJECTIVE To compare the orthogonal diameter (visual metric) method against a manual perimeter tracing (planimetry) method to measure volume of brain tumors in dogs by use of MRI scans. SAMPLE 22 sets of MRI brain scans pertaining to 22 client-owned dogs with histologically confirmed glioma. PROCEDURES MRI scans were reviewed by 2 operators, and scans revealing tumors with a degree of gadolinium enhancement that allowed discrimination between tumor tissue and healthy parenchyma were used. Each operator calculated tumor volume for each set of scans twice by use of visual metric and planimetry methods. Inter- and intraoperator variability were assessed by calculation of an agreement index (AI). RESULTS Mean ± SD intraoperator AIs were 0.79 ± 0.24 for the visual metric method and 0.89 ± 0.17 for the planimetry method. Intraoperator variability for both operators was significantly less when the planimetry method was used than when the visual metric method was used. No significant differences were identified in mean interoperator AI between visual metric (0.68 ± 0.28) and planimetry (0.67 ± 0.31) methods. CONCLUSIONS AND CLINICAL RELEVANCE The lower intraoperator variability achieved with the planimetry versus visual metric method should result in more precise volume assessments when the same operator performs multiple volume measurements of brain tumors in dogs. Equivocal results for interoperator variability may have been due to method variance or inadequate preliminary training. Additional studies are needed to evaluate the suitability of planimetry for assessing response to treatment.

Glioblastoma Segmentation: Comparison of Three Different Software Packages

To facilitate a more widespread use of volumetric tumor segmentation in clinical studies, there is an urgent need for reliable, user-friendly segmentation software. The aim of this study was therefore to compare three different software packages for semi-automatic brain tumor segmentation of glioblastoma; namely BrainVoyager^TM QX, ITK-Snap and 3D Slicer, and to make data available for future reference. Pre-operative, contrast enhanced T₁-weighted 1.5 or 3 Tesla Magnetic Resonance Imaging (MRI) scans were obtained in 20 consecutive patients who underwent surgery for glioblastoma. MRI scans were segmented twice in each software package by two investigators. Intra-rater, inter-rater and between-software agreement was compared by using differences of means with 95% limits of agreement (LoA), Dice’s similarity coefficients (DSC) and Hausdorff distance (HD). Time expenditure of segmentations was measured using a stopwatch. Eighteen tumors were included in the analyses. Inter-rater agreement was highest for BrainVoyager with difference of means of 0.19 mL and 95% LoA from -2.42 mL to 2.81 mL. Between-software agreement and 95% LoA were very similar for the different software packages. Intra-rater, inter-rater and between-software DSC were ≥ 0.93 in all analyses. Time expenditure was approximately 41 min per segmentation in BrainVoyager, and 18 min per segmentation in both 3D Slicer and ITK-Snap. Our main findings were that there is a high agreement within and between the software packages in terms of small intra-rater, inter-rater and between-software differences of means and high Dice’s similarity coefficients. Time expenditure was highest for BrainVoyager, but all software packages were relatively time-consuming, which may limit usability in an everyday clinical setting.

Quantitative radiology: applications to oncology

Oncologists, clinician-scientists, and basic scientists collect computed tomography, magnetic resonance, and positron emission tomography images in the process of caring for patients, managing clinical trials, and investigating cancer biology. As we have developed more sophisticated means for noninvasively delineating and characterizing neoplasms, these image data have come to play a central role in oncology. In parallel, the increasing complexity and volume of these data have necessitated the development of quantitative methods for assessing tumor burden, and by proxy, disease-free survival.

Accelerating Fuzzy-C Means Using an Estimated Subsample Size

Many algorithms designed to accelerate the Fuzzy c-Means (FCM) clustering algorithm randomly sample the data. Typically, no statistical method is used to estimate the subsample size, despite the impact subsample sizes have on speed and quality. This paper introduces two new accelerated algorithms, GOFCM and MSERFCM, that use a statistical method to estimate the subsample size. GOFCM, a variant of SPFCM, also leverages progressive sampling. MSERFCM, a variant of rseFCM, gains a speedup from improved initialization. A general, novel stopping criterion for accelerated clustering is introduced. The new algorithms are compared to FCM and four accelerated variants of FCM. GOFCM's speedup was 4-47 times that of FCM and faster than SPFCM on each of the six datasets used in experiments. For five of the datasets, partitions were within 1% of those of FCM. MSERFCM's speedup was 5-26 times that of FCM and produced partitions within 3% of those of FCM on all datasets. A unique dataset, consisting of plankton images, exposed the strengths and weaknesses of many of the algorithms tested. It is shown that the new stopping criterion is effective in speeding up algorithms such as SPFCM and the final partitions are very close to those of FCM.

Computerized brain tumor segmentation in magnetic resonance imaging

OBJECTIVE

To propose an automatic brain tumor segmentation system.

METHODS

The system used texture characteristics as its main source of information for segmentation.

RESULTS

The mean correct match was 94% of correspondence between the segmented areas and ground truth.

CONCLUSION

Final results showed that the proposed system was able to find and delimit tumor areas without requiring any user interaction.

Reproducibility of scan prescription in follow-up brain MRI: manual versus automatic determination

In follow-up brain magnetic resonance imaging (MRI), precise reproducibility of the scan prescription is important so that over- or underestimating changes in volumes of clinical interest is prevented. (The scan prescription is defined as the location and orientation of the head with respect to the scan planes of the three-dimensional MRI matrix.) In this study, the misregistration between the original and a second scan was calculated in the case of both manual positioning and automatic positioning. These calculations were carried out both for a healthy volunteer scanned repeatedly and, in a retrospective study, for 225 patients who had an original and at least one follow-up scan. The effects of the scan operator being the same for both scans or being different were also examined. A commercially available 1.5 Tesla MRI system and a six-element head-array coil were employed in all of the imaging. The reproducibility of the scan prescription was determined by the registration of the original scan image to the follow-up scan image by use of the Fourier phase correlation method. Our results showed that (1) the reproducibility by automatic positioning was superior to that by manual positioning (p < 0.05), and (2) there was no significant difference in the results between when the operator was the same or different (p > 0.05). We conclude that, in follow-up brain MRI, automatic positioning should be used, because manual positioning decreases the reproducibility of the scan prescription even if the same operator performs the second scan.

Validation techniques for quantitative brain tumors measurements

Quantitative measurements of tumor volume becomes more realistic with the use of imaging- particularly specially when the tumor have non-ellipsoidal morphology, which remains subtle, irregular and difficult to assess by visual metric and clinical examination. The quantitative measurements depend strongly on the accuracy of the segmentation technique. The validity of brain tumor segmentation methods is an important issue in medical imaging because it has a direct impact on many applications such as surgical planning and quantitative measurements of tumor volume. Our goal was to examine two popular segmentation techniques seeded region growing and active contour "snakes" to be compared against experts' manual segmentations as the gold standard. We illustrated these methods on brain tumor volume cases using MR imaging modality.

Automatic segmentation of meningioma from non-contrasted brain MRI integrating fuzzy clustering and region growing

Background

In recent years, magnetic resonance imaging (MRI) has become important in brain tumor diagnosis. Using this modality, physicians can locate specific pathologies by analyzing differences in tissue character presented in different types of MR images.

This paper uses an algorithm integrating fuzzy-c-mean (FCM) and region growing techniques for automated tumor image segmentation from patients with menigioma. Only non-contrasted T1 and T2 -weighted MR images are included in the analysis. The study's aims are to correctly locate tumors in the images, and to detect those situated in the midline position of the brain.

Methods

The study used non-contrasted T1- and T2-weighted MR images from 29 patients with menigioma. After FCM clustering, 32 groups of images from each patient group were put through the region-growing procedure for pixels aggregation. Later, using knowledge-based information, the system selected tumor-containing images from these groups and merged them into one tumor image. An alternative semi-supervised method was added at this stage for comparison with the automatic method. Finally, the tumor image was optimized by a morphology operator. Results from automatic segmentation were compared to the "ground truth" (GT) on a pixel level. Overall data were then evaluated using a quantified system.

Results

The quantified parameters, including the "percent match" (PM) and "correlation ratio" (CR), suggested a high match between GT and the present study's system, as well as a fair level of correspondence. The results were compatible with those from other related studies. The system successfully detected all of the tumors situated at the midline of brain.

Six cases failed in the automatic group. One also failed in the semi-supervised alternative. The remaining five cases presented noticeable edema inside the brain. In the 23 successful cases, the PM and CR values in the two groups were highly related.

Conclusions

Results indicated that, even when using only two sets of non-contrasted MR images, the system is a reliable and efficient method of brain-tumor detection. With further development the system demonstrates high potential for practical clinical use.

MRI internal segmentation of optic pathway gliomas: clinical implementation of a novel algorithm

PURPOSE

Optic pathway gliomas (OPGs) are diagnosed based on typical MR features and require careful monitoring with serial MRI. Reliable, serial radiological comparison of OPGs is a difficult task, where accuracy becomes very important for clinical decisions on treatment initiation and results. Current radiological methodology usually includes linear measurements that are limited in terms of precision and reproducibility.

METHOD

We present a method that enables semiautomated segmentation and internal classification of OPGs using a novel algorithm. Our method begins with co-registration of the different sequences of an MR study so that T1 and T2 slices are realigned. The follow-up studies are then re-sliced according to the baseline study. The baseline tumor is segmented, with internal components classified into solid non-enhancing, solid-enhancing, and cystic components, and the volume is calculated. Tumor demarcation is then transferred onto the next study and the process repeated. Numerical values are correlated with clinical data such as treatment and visual ability.

RESULTS

We have retrospectively implemented our method on 24 MR studies of three OPG patients. Clinical case reviews are presented here. The volumetric results have been correlated with clinical data and their implications are also discussed.

CONCLUSIONS

The heterogeneity of OPGs, the long course, and the young age of the patients are all driving the demand for more efficient and accurate means of tumor follow-up. This method may allow better understanding of the natural history of the tumor and provide a more advanced means of treatment evaluation.

Monitoring radiographic brain tumor progression

Determining radiographic progression in primary malignant brain tumors has posed a significant challenge to the neuroncology community. Glioblastoma multiforme (GBM, WHO Grade IV) through its inherent heterogeneous enhancement, growth patterns, and irregular nature has been difficult to assess for progression. Our ability to detect tumor progression radiographically remains inadequate. Despite the advanced imaging techniques, detecting tumor progression continues to be a clinical challenge. Here we review the different criteria used to detect tumor progression, and highlight the inherent challenges with detection of progression.

A novel method for volumetric MRI response assessment of enhancing brain tumors

Current radiographic response criteria for brain tumors have difficulty describing changes surrounding postoperative resection cavities. Volumetric techniques may offer improved assessment, however usually are time-consuming, subjective and require expert opinion and specialized magnetic resonance imaging (MRI) sequences. We describe the application of a novel volumetric software algorithm that is nearly fully automated and uses standard T1 pre- and post-contrast MRI sequences. T1-weighted pre- and post-contrast images are automatically fused and normalized. The tumor region of interest is grossly outlined by the user. An atlas of the nasal mucosa is automatically detected and used to normalize levels of enhancement. The volume of enhancing tumor is then automatically calculated. We tested the ability of our method to calculate enhancing tumor volume with resection cavity collapse and when the enhancing tumor is obscured by subacute blood in a resection cavity. To determine variability in results, we compared narrowly-defined tumor regions with tumor regions that include adjacent meningeal enhancement and also compared different contrast enhancement threshold levels used for the automatic calculation of enhancing tumor volume. Our method quantified enhancing tumor volume despite resection cavity collapse. It detected tumor volume increase in the midst of blood products that incorrectly caused decreased measurements by other techniques. Similar trends in volume changes across scans were seen with inclusion or exclusion of meningeal enhancement and despite different automated thresholds for tissue enhancement. Our approach appears to overcome many of the challenges with response assessment of enhancing brain tumors and warrants further examination and validation.

[Chemotherapy and targeted treatments in glioblastomas]

BACKGROUND AND PURPOSE

We review the indications and limitations of chemotherapy in glioblastoma multiform (GBM), including the role played by alkylating and other cytotoxic agents and the increased input of clinical research on targeted agents in GBM management.

METHODS

In 2005, a phase III study clearly concluded in the benefit of adding temozolomide during and after radiotherapy treatment in GBM and thus defined the new standard of treatment in this devastating disease. This schedule increased the median survival from 12.1 to 14 months and the two- and five-year survival rates from 8 to 26%, and 3 to 10%, respectively, with a good tolerance profile. Moreover, methylation of the promoter of the O6 methylguanine DNA transferase (MGMT) gene exhibits a prognostic impact independently of therapeutic schedule but may also predict the benefit of adding temozolomide to radiotherapy. However, pitfalls in MGMT determination and lack of prospective validation have to be solved before considering MGMT as a decisional marker. More recently, antiangiogenic agents including enzastaurin, cediranib, bevacizumab, and others that target mainly the VEGF pathway, have been evaluated in this highly angiogenic disease. Among them, only bevacizumab has been associated with clear anti-tumor activity, although the lack of control studies limits the impact of the results to date.

CONCLUSIONS

Recent studies conducted in GBM, both in the adjuvant and recurrent setting, have changed the natural course of the disease and opened a new area of clinical research, including imaging and response evaluation, predictive markers, and other targeted therapies.

Change Detection & Characterization: A New Tool for Imaging Informatics and Cancer Research

Modern imaging systems are able to produce a rich and diverse array of information, regarding various facets of anatomy and function. The quantity of information produced by these systems is so bountiful, however, as to have the potential to become a hindrance to clinical assessment. In the context of serial image evaluation, computer-based change detection and characterization is one important mechanism to process the information produced by imaging systems, so as to reduce the quantity of data, direct the attention of the physician to regions of the data which are the most informative for their purposes, and present the data in the form in which it will be the most useful. Change detection and characterization algorithms may serve as a basis for the creation of an objective definition of progression, which will reduce inter and intra-observer variability, and facilitate earlier detection of disease and recurrence, which in turn may lead to improved outcomes. Decreased observer variability combined with increased acuity should make it easier to discover promising therapies. Quantitative measures of the response to these therapies should provide a means to compare the effectiveness of treatments under investigation. Change detection may be applicable to a broad range of cancers, in essentially all anatomical regions. The source of information upon which change detection comparisons may be based is likewise broad. Validation of algorithms for the longitudinal assessment of cancer patients is expected to be challenging, though not insurmountable, as the many facets of the problem mean that validation will likely need to be approached from a variety of vantage points. Change detection and characterization is quickly becoming a very active field of investigation, and it is expected that this burgeoning field will help to facilitate cancer care both in the clinic and research.

[Cerebral gliomas: imaging diagnosis and follow-up]

The management of gliomas evolves towards more aggressive strategies with a combination of surgery, radiotherapy and chemotherapy. Follow-up imaging based on morphological MRI, with simple and reproducible protocols, may be associated with functional MRI and spectroscopy. Baseline postsurgical MRI must be performed within the first three days. Follow-up examinations should be done 2 months after radiotherapy and during chemotherapy, usually after each cycle of two or three treatments. Continued follow-up after therapy is recommended to assess response and detect recurrences or therapeutic complications.

Comparison of manual and automatic section positioning of brain MR images

The study protocol was approved by the institutional review board and was in full compliance with HIPAA guidelines. Informed consent was obtained from all patients. The purpose of this study was to prospectively compare intra- and intersubject variability of manual versus automatic magnetic resonance (MR) imaging section prescription. In two examinations, T2-weighted series were acquired with both methods. All intrasubject and three of six intersubject section prescription variances were significantly higher for manual prescription (P < .01). Root mean square errors confirmed better coregistration of the automated approach (P < .001). Automatic section prescription leads to improved reproducibility of imaging orientations for intra- and intersubject series in clinical practice.

CT assessment of the effect of fluticasone propionate aqueous nasal spray treatment on lower turbinate hypertrophy due to vasomotor rhinitis

CONCLUSION

Fluticasone propionate (FP) aqueous nasal spray was objectively found to be effective and safe for the treatment of lower turbinate enlargement in patients with vasomotor rhinitis.

OBJECTIVE

To assess the efficacy of FP aqueous nasal spray treatment in lower turbinate hypertrophy due to vasomotor rhinitis using CT.

MATERIAL AND METHODS

Of 35 patients with hypertrophic lower turbinates due to vasomotor rhinitis, 20 were treated twice daily with FP aqueous nasal spray (200 microg/day) for 3 months continuously and 15 were treated with placebo vehicle as a control group. The local effect of the nasal spray was studied using CT and visual analog scales.

RESULTS

Treatment with FP provided significantly greater relief from the symptom of nasal obstruction compared with placebo over the entire 3-month treatment period (p < 0.001). When the change from baseline was compared between the two groups, FP produced statistically significant reductions in the mucosal area of the lower turbinates and in the thickness of the nasal mucosa after 3 months (p < 0.05).

A soft-segmentation visualization scheme for magnetic resonance images

Prevalent visualization tools exploit gray value distribution in images through modified histogram equalization and matching technique, referred to as the window width/window level-based method, to improve visibility and enhance diagnostic value. The window width/window level tool is extensively used in magnetic resonance (MR) images to highlight tissue boundaries during image interpretation. However, the identification of different regions and distinct boundaries between them based on gray-level distribution and displayed intensity levels is extremely difficult because of the large dynamic range of tissue intensities inherent in MR images. We propose a soft-segmentation visualization scheme to generate pixel partitions from the histogram of MR image data using a connectionist approach and then generate selective visual depictions of pixel partitions using pseudo color based on an appropriate fuzzy membership function. By applying the display scheme in clinical examples in this study, we could demonstrate additional overlapping regions between distinct tissue types in healthy and diseased areas (in the brain) that could help improve the tissue characterization ability of MR images.

Segmentation of tumors in magnetic resonance brain images using an interactive multiscale watershed algorithm

RATIONALE AND OBJECTIVE

This article presents the evaluation of an interactive multiscale watershed segmentation algorithm for segmenting tumors in magnetic resonance brain images of patients scheduled for neuronavigational procedures.

MATERIALS AND METHODS

The watershed method is compared with manual delineation with respect to accuracy, repeatability, and efficiency.

RESULTS

In the 20 patients included in this study, the measured volume of the tumors ranged from 2.7 to 81.9 cm(3). A comparison of the tumor volumes measured with watershed segmentation to the volumes measured with manual delineation shows that the two methods are interchangeable according to the Bland and Altman criterion, and thus equally accurate. The repeatability of the watershed method and the manual method are compared by looking at the similarity of the segmented volumes. The similarity for intraobserver and interobserver variability for watershed segmentation is 96.4% and 95.3%, respectively, compared with 93.5% and 90.0% for manual outlining, from which it may be concluded that the watershed method is more repeatable. Moreover, the watershed algorithm is on average three times faster than manual outlining.

CONCLUSION

The watershed method has an accuracy comparable to that of manual delineation and outperforms manual outlining on the criteria of repeatability and efficiency.

Avaliação textural por ressonância magnética dos tumores da fossa posterior em crianças

OBJETIVO: Possibilitar a distinção entre tecidos sãos e patológicos em pacientes da faixa etária pediátrica portadores de tumores da fossa posterior, por meio da análise de parâmetros texturais calculados a partir de imagens de ressonância magnética. MATERIAIS E MÉTODOS: Foram analisados 14 pacientes da faixa etária pediátrica, portadores de tumores da fossa posterior, através da definição dos valores texturais das regiões de interesse representando tecidos sãos e patológicos, com base em imagens de ressonância magnética pesadas em T2 pelo "software" MaZda. RESULTADOS: Houve diferença estatisticamente significativa entre os tecidos normal e tumoral, bem como entre os tecidos presumidamente normais adjacentes e distantes da lesão. Não foi possível a distinção entre edema e tumor. CONCLUSÃO: A avaliação textural por ressonância magnética é uma técnica útil para a determinação de diferenças entre diversos tipos de tecidos, inclusive entre áreas de tecidos presumidamente normais à análise visual.

A review of the automated detection of change in serial imaging studies of the brain

Serial imaging is frequently performed on patients with diseases of the brain, to track and observe changes. Magnetic resonance imaging provides very detailed and rich information, and is therefore used frequently for this application. The data provided by MR can be so plentiful; however, that it obfuscates the information the radiologist seeks. A system which could reduce the large quantity of primitive data to a smaller and more informative subset of data, emphasizing change, would be useful. This article discusses motivating factors for the production of an automated process to this effect, and reviews the approaches of previous authors. The discussion is focused on brain tumors and multiple sclerosis, but many of the ideas are applicable to other disease processes, as well.

Brain tumor target volume determination for radiation treatment planning through automated MRI segmentation

PURPOSE

To assess the effectiveness of two automated magnetic resonance imaging (MRI) segmentation methods in determining the gross tumor volume (GTV) of brain tumors for use in radiation therapy treatment planning.

METHODS AND MATERIALS

Two automated MRI tumor segmentation methods (supervised k-nearest neighbors [kNN] and automatic knowledge-guided [KG]) were evaluated for their potential as "cyber colleagues." This required an initial determination of the accuracy and variability of radiation oncologists engaged in the manual definition of the GTV in MRI registered with computed tomography images for 11 glioma patients. Three sets of contours were defined for each of these patients by three radiation oncologists. These outlines were compared directly to establish inter- and intraoperator variability among the radiation oncologists. A novel, probabilistic measurement of accuracy was introduced to compare the level of agreement among the automated MRI segmentations. The accuracy was determined by comparing the volumes obtained by the automated segmentation methods with the weighted average volumes prepared by the radiation oncologists.

RESULTS

Intra- and inter-operator variability in outlining was found to be an average of 20% +/- 15% and 28% +/- 12%, respectively. Lowest intraoperator variability was found for the physician who spent the most time producing the contours. The average accuracy of the kNN segmentation method was 56% +/- 6% for all 11 cases, whereas that of the KG method was 52% +/- 7% for 7 of the 11 cases when compared with the physician contours. For the areas of the contours where the oncologists were in substantial agreement (i.e., the center of the tumor volume), the accuracy of kNN and KG was 75% and 72%, respectively. The automated segmentation methods were found to be least accurate in outlining at the edges of the tumor volume.

CONCLUSIONS

The kNN method was able to segment all cases, whereas the KG method was limited to enhancing tumors and gliomas with clear enhancing edges and no cystic formation. Both methods undersegment the tumor volume when compared with the radiation oncologists and performed within the variability of the contouring performed by experienced radiation oncologists based on the same data.

Quantification of tumour response to radiotherapy

In 1979, the World Health Organization (WHO) established criteria based on tumour volume change for classifying response to therapy as (i) progressive disease (PD), (ii) partial recovery (PR), and (iii) no change (NC). Typically, the tumour volume is reported from diameter measurements, using the calliper method. Alternatively, the Cavalieri method provides unbiased volume estimates of any structure without assumptions about its shape. In this study, we applied the Cavalieri method in combination with point counting to investigate the changes in tumour volume in four patients with high grade glioma, using 3D MRI. In particular, the volume of tumour within the enhancement boundary, the enhancing abnormality (EA), was estimated from T(1) weighted images, and the volume of the non-enhancing abnormality, (NEA) enhancing abnormality, was estimated from T(2) relaxation time and magnetic transfer ratio tissue characterization maps. We compared changes in tumour volume estimated by the Cavalieri method with those obtained using the calliper method. Absolute tumour volume differed significantly between the two methods. Analysis of relative change in tumour volume, based on the WHO criteria, provided a different classification using the calliper and Cavalieri methods. The benefit of the Cavalieri method over the calliper method in the estimation of tumour volume is justified by the following factors. First, Cavalieri volume estimates are mathematically unbiased. Second, the Cavalieri method is highly efficient under an appropriate sampling density (i.e. EA volume estimates can be obtained with a coefficient of error no higher than 5% in 2-3 min). Third, the source of variation of the volume estimates due to disagreements between observers, and within observer, is much greater in the positioning of the calliper diameters than in the identification of the tumour boundaries when applying the Cavalieri method. Additionally, the error prediction formula, available to estimate the coefficient of error of Cavalieri volume estimates from the data, allows us to establish more precise classification criteria against which to identify potentially clinical significant changes in tumour volume.

Angiogenesis in glioma: molecular mechanisms and roadblocks to translation

Gliomas are characterized by very high levels of neo-vascularization holding out the hope that therapies aimed at angiogenesis will have a significant impact on this intractable family of tumors. Intense research into the molecular mechanisms that drive the formation of new blood vessels in response to tumor growth has revealed a great deal of complexity, at the heart of which are competing pro- and anti-angiogenic influences. The relevant signaling pathways, and how they might be manipulated to interfere in the promotion of vessel growth are discussed. Several types of anti-angiogenic lead compounds are already in clinical trials, but assessing their impact on brain tumors is not straightforward. We discuss in depth some of the practical aspects of using imaging to more meaningfully follow tumor progression and response to treatment, which is particularly relevant to the use of therapies that target blood flow directly, which is fundamental to modern imaging modalities.

Multispectral magnetic resonance image analysis using principal component and linear discriminant analysis

PURPOSE

To explore the possibilities of combining multispectral magnetic resonance (MR) images of different patients within one data matrix.

MATERIALS AND METHODS

Principal component and linear discriminant analysis was applied to multispectral MR images of 12 patients with different brain tumors. Each multispectral image consisted of T1-weighted, T2-weighted, proton-density-weighted, and gadolinium-enhanced T1-weighted MR images, and a calculated relative regional cerebral blood volume map.

RESULTS

Similar multispectral image regions were clustered, while dissimilar multispectral image regions were scattered in a single plot. Both principal component and linear discriminant analysis allowed discrimination between healthy and tumor regions on the image. In addition, linear discriminant analysis allowed discrimination between oligodendrogliomas and astrocytomas. However, the discriminant analysis method was partially capable of recognizing the tumor identity in unknown multispectral images.

CONCLUSION

The proposed method may help the radiologist in comparing multispectral MR images of different patients in a more easy and objective way.

Improving image contrast using principal component analysis for subsequent image segmentation

This article presents a technique for improving MR image contrast by linearly combining multiple MR images with different tissue contrast. The weighting coefficients of the linear combination are derived using principal component analysis. The contrast-enhanced composite image is segmented subsequently using gray level-based 1D segmentation methods. The technique reduces a multispectral image set to composite eigenimages and allows application of appropriate 1D segmentation methods that do not have equivalent counterparts in multispectral methods.

Mapping therapeutic response in a patient with malignant glioma

Short-interval scanning of patients offers a detailed understanding of the natural progression of tumor tissue, as revealed through imaging markers such as contrast enhancement and edema, prior to therapy. Following treatment, short-interval scanning can also provide evidence of attenuation of growth rates. We present a longitudinal imaging study of a patient with glioblastoma multiforme (GBM) scanned 15 times in 104 days on a 3 T MR scanner. Images were analyzed independently by two automated algorithms capable of creating detailed maps of tumor changes as well as volumetric analysis. The algorithms, a nearest-neighbor-based tissue segmentation and a surface-modeling algorithm, tracked the patient's response to temozolomide, showing an attenuation of growth. The need for surrogate imaging end-points, of which growth rates are an example, is discussed. Further, the strengths of these algorithms, the insight gained by short-interval scanning, and the need for a better understanding of imaging markers are also described.

Results of a pilot study involving the use of an antisense oligodeoxynucleotide directed against the insulin-like growth factor type I receptor in malignant astrocytomas

PURPOSE

Preclinical animal experiments support the use of an antisense oligodeoxynucleotide directed against the insulin-like growth factor type I receptor (IGF-IR/AS ODN) as an effective potential antitumor agent. We performed a human pilot safety and feasibility study using an IGF-IR/AS ODN strategy in patients with malignant astrocytoma.

PATIENTS AND METHODS

Autologous glioma cells collected at surgery were treated ex vivo with an IGF-IR/AS ODN, encapsulated in diffusion chambers, reimplanted in the rectus sheath within 24 hours of craniotomy, and retrieved after a 24-hour in situ incubation. Serial posttreatment assessments included clinical examination, laboratory studies, and magnetic resonance imaging scans.

RESULTS

Other than deep venous thrombosis noted in some patients, no other treatment-related side effects were observed. IGF-IR/AS ODN-treated cells, when retrieved and assessed, were < or = 2% intact by trypan blue exclusion, and none of the intact cells were viable in culture thereafter. Parallel Western blots disclosed IGF-IR downregulation to < or = 10% after ex vivo antisense treatment. At follow-up, clinical and radiographic improvements were observed in eight of 12 patients, including three cases of distal recurrence with unexpected spontaneous or postsurgical regression at either the primary or the distant intracranial site.

CONCLUSION

Ex vivo IGF-IR/AS ODN treatment of autologous glioma cells induces apoptosis and a host response in vivo without unusual side effects. Subsequent transient and sustained radiographic and clinical improvements warrant further clinical investigations.

Comparison of diameter and perimeter methods for tumor volume calculation

PURPOSE

Lesion volume is often used as an end point in clinical trials of oncology therapy. We sought to compare the common method of using orthogonal diameters to estimate lesion volume (the diameter method) with a computer-assisted planimetric technique (the perimeter method).

METHODS

Radiologists reviewed 825 magnetic resonance imaging studies from 219 patients with glioblastoma multiforme. Each study had lesion volume independently estimated via the diameter and perimeter methods. Cystic areas were subtracted out or excluded from the outlined lesion. Inter- and intrareader variability was measured by using multiple readings on 48 cases. Where serial studies were available in noncystic cases, a mock response analysis was used.

RESULTS

The perimeter method had a reduced interreader and intrareader variability compared with the diameter method (using SD of differences): intrareader, 1.76 mL v 7.38 mL (P < .001); interreader, 2.51 mL v 9.07 mL (P < .001) for perimeter and diameter results, respectively. Of the 121 noncystic cases, 23 had serial data. In six (26.1%) of those 23, a classification difference occurred when the perimeter method was used versus the diameter method.

CONCLUSION

Variability of measurements was reduced with the computer-assisted perimeter method compared with the diameter method, which suggests that changes in volume can be detected more accurately with the perimeter method. The differences between these techniques seem large enough to have an impact on grading the response to therapy.

Automatic segmentation of non-enhancing brain tumors in magnetic resonance images

Tumor segmentation from magnetic resonance (MR) images may aid in tumor treatment by tracking the progress of tumor growth and/or shrinkage. In this paper we present the first automatic segmentation method which separates non-enhancing brain tumors from healthy tissues in MR images to aid in the task of tracking tumor size over time. The MR feature images used for the segmentation consist of three weighted images (T1, T2 and proton density (PD)) for each axial slice through the head. An initial segmentation is computed using an unsupervised fuzzy clustering algorithm. Then, integrated domain knowledge and image processing techniques contribute to the final tumor segmentation. They are applied under the control of a knowledge-based system. The system knowledge was acquired by training on two patient volumes (14 images). Testing has shown successful tumor segmentations on four patient volumes (31 images). Our results show that we detected all six non-enhancing brain tumors, located tumor tissue in 35 of the 36 ground truth (radiologist labeled) slices containing tumor and successfully separated tumor regions from physically connected CSF regions in all the nine slices. Quantitative measurements are promising as correspondence ratios between ground truth and segmented tumor regions ranged between 0.368 and 0.871 per volume, with percent match ranging between 0.530 and 0.909 per volume.

Determination of intracranial tumor volumes in a rodent brain using magnetic resonance imaging, Evans blue, and histology: a comparative study

The measurement of tumor volumes is a practical and objective method of assessing the efficacy of a therapeutic agent. However, the relative accuracy of different methods of assessing tumor volume has been unclear. Using T1-weighted, gadolinium-enhanced magnetic resonance Imaging (T1-MRI), Evans Blue infusion and histology we measured intracranial tumor volumes in a rodent brain tumor model (RT2) at days 10, 16 and 18 after implantation of cells in the caudate putamen. There is a good correlation between tumor volumes comparing T1-MRI and Evans Blue (r2 = 0.99), T1-MRI and Histology (r2 = 0.98) and histology and Evans Blue (r2 = 0.93). Each of these methods is reliable in estimating tumor volumes in laboratory animals. There was significant uptake of gadolinium and Evans Blue in the tumor suggesting a wide disruption of the blood-brain barrier.

Volumetric reduction of a choroid plexus carcinoma using preoperative chemotherapy

We report for the first time a measured volumetric reduction of a choroid plexus carcinoma utilizing preoperative chemotherapy. Histologically proven choroid plexus carcinoma was diagnosed in a fifteen month old female. She was treated with three courses of chemotherapy including etoposide (VP16), cyclophosphamide, vincristine, and cisplatin. Computer-assisted three dimensional reconstruction of the tumor volume was performed prior to and after three courses of chemotherapy. An overall reduction of 29.5% of tumor volume was accomplished preoperatively. Staged surgical procedures resulted in a complete resection of her lesion and she has remained disease-free for 31 months. A volumetric measurement as a response to preoperative chemotherapy may prove valuable in determining future optimal treatment regimens for choroid plexus carcinoma of childhood.