Distinguishing recurrence from radiation-induced lung injury at the time of RECIST progressive disease on post-SABR CT scans using radiomics

Stereotactic ablative radiotherapy (SABR) is a highly effective treatment for patients with early-stage lung cancer who are inoperable. However, SABR causes benign radiation-induced lung injury (RILI) which appears as lesion growth on follow-up CT scans. This triggers the standard definition of progressive disease, yet cancer recurrence is not usually present, and distinguishing RILI from recurrence when a lesion appears to grow in size is critical but challenging. In this study, we developed a tool to do this using scans with apparent lesion growth after SABR from 68 patients. We performed bootstrapped experiments using radiomics and explored the use of multiple regions of interest (ROIs). The best model had an area under the receiver operating characteristic curve of 0.66 and used a sphere with a diameter equal to the lesion's longest axial measurement as the ROI. We also investigated the effect of using inter-feature and volume correlation filters and found that the former was detrimental to performance and that the latter had no effect. We also found that the radiomics features ranked as highly important by the model were significantly correlated with outcomes. These findings represent a key step in developing a tool that can help determine who would benefit from follow-up invasive interventions when a SABR-treated lesion increases in size, which could help provide better treatment for patients.

A Visual Aid Tool for Detection of Pancreatic Tumour-Vessel Contact on Staging CT: A Retrospective Cohort Study

Purpose: In pancreatic adenocarcinoma, the difficult distinction between normal and affected pancreas on CT studies may lead to discordance between the pre-surgical assessment of vessel involvement and intraoperative findings. We hypothesize that a visual aid tool could improve the performance of radiology residents when detecting vascular invasion in pancreatic adenocarcinoma patients. Methods: This study consisted of 94 pancreatic adenocarcinoma patient CTs. The visual aid compared the estimated body fat density of each patient with the densities surrounding the superior mesenteric artery and mapped them onto the CT scan. Four radiology residents annotated the locations of perceived vascular invasion on each scan with the visual aid overlaid on alternating scans. Using 3 expert radiologists as the reference standard, we quantified the area under the receiver operating characteristic curve to determine the performance of the tool. We then used sensitivity, specificity, balanced accuracy ((sensitivity + specificity)/2), and spatial metrics to determine the performance of the residents with and without the tool. Results: The mean area under the curve was 0.80. Radiology residents' sensitivity/specificity/balanced accuracy for predicting vascular invasion were 50%/85%/68% without the tool and 81%/79%/80% with it compared to expert radiologists, and 58%/85%/72% without the tool and 78%/77%/77% with it compared to the surgical pathology. The tool was not found to impact the spatial metrics calculated on the resident annotations of vascular invasion. Conclusion: The improvements provided by the visual aid were predominantly reflected by increased sensitivity and accuracy, indicating the potential of this tool as a learning aid for trainees.

Predicting stereotactic radiosurgery outcomes with multi-observer qualitative appearance labelling versus MRI radiomics

Qualitative observer-based and quantitative radiomics-based analyses of T1w contrast-enhanced magnetic resonance imaging (T1w-CE MRI) have both been shown to predict the outcomes of brain metastasis (BM) stereotactic radiosurgery (SRS). Comparison of these methods and interpretation of radiomics-based machine learning (ML) models remains limited. To address this need, we collected a dataset of n = 123 BMs from 99 patients including 12 clinical features, 107 pre-treatment T1w-CE MRI radiomic features, and BM post-SRS progression scores. A previously published outcome model using SRS dose prescription and five-way BM qualitative appearance scoring was evaluated. We found high qualitative scoring interobserver variability across five observers that negatively impacted the model's risk stratification. Radiomics-based ML models trained to replicate the qualitative scoring did so with high accuracy (bootstrap-corrected AUC = 0.84-0.94), but risk stratification using these replicated qualitative scores remained poor. Radiomics-based ML models trained to directly predict post-SRS progression offered enhanced risk stratification (Kaplan-Meier rank-sum p = 0.0003) compared to using qualitative appearance. The qualitative appearance scoring enabled interpretation of the progression radiomics-based ML model, with necrotic BMs and a subset of heterogeneous BMs predicted as being at high-risk of post-SRS progression, in agreement with current radiobiological understanding. Our study's results show that while radiomics-based SRS outcome models out-perform qualitative appearance analysis, qualitative appearance still provides critical insight into ML model operation.

Regulation of Skeletal Muscle Resistance Arteriolar Tone: Integration of Multiple Mechanisms

INTRODUCTION

Physiological system complexity represents an imposing challenge to gaining insight into how arteriolar behavior emerges. Further, mechanistic complexity in arteriolar tone regulation requires that a systematic determination of how these processes interact to alter vascular diameter be undertaken.

METHODS

The present study evaluated the reactivity of ex vivo proximal and in situ distal resistance arterioles in skeletal muscle with challenges across the full range of multiple physiologically relevant stimuli and determined the stability of responses over progressive alterations to each other parameter. The five parameters chosen for examination were (1) metabolism (adenosine concentration), (2) adrenergic activation (norepinephrine concentration), (3) myogenic activation (intravascular pressure), (4) oxygen (superfusate PO2), and (5) wall shear rate (altered intraluminal flow). Vasomotor tone of both arteriole groups following challenge with individual parameters was determined; subsequently, responses were determined following all two- and three-parameter combinations to gain deeper insight into how stimuli integrate to change arteriolar tone. A hierarchical ranking of stimulus significance for establishing arteriolar tone was performed using mathematical and statistical analyses in conjunction with machine learning methods.

RESULTS

Results were consistent across methods and indicated that metabolic and adrenergic influences were most robust and stable across all conditions. While the other parameters individually impact arteriolar tone, their impact can be readily overridden by the two dominant contributors.

CONCLUSION

These data suggest that mechanisms regulating arteriolar tone are strongly affected by acute changes to the local environment and that ongoing investigation into how microvessels integrate stimuli regulating tone will provide a more thorough understanding of arteriolar behavior emergence across physiological and pathological states.

Application of a novel index for understanding vascular health following pharmacological intervention in a pre-clinical model of metabolic disease

While a thorough understanding of microvascular function in health and how it becomes compromised with progression of disease risk is critical for developing effective therapeutic interventions, our ability to accurately assess the beneficial impact of pharmacological interventions to improve outcomes is vital. Here we introduce a novel Vascular Health Index (VHI) that allows for simultaneous assessment of changes to vascular reactivity/endothelial function, vascular wall mechanics and microvessel density within cerebral and skeletal muscle vascular networks with progression of metabolic disease in obese Zucker rats (OZR); under control conditions and following pharmacological interventions of clinical relevance. Outcomes are compared to "healthy" conditions in lean Zucker rats. We detail the calculation of vascular health index, full assessments of validity, and describe progressive changes to vascular health index over the development of metabolic disease in obese Zucker rats. Further, we detail the improvement to cerebral and skeletal muscle vascular health index following chronic treatment of obese Zucker rats with anti-hypertensive (15%-52% for skeletal muscle vascular health index; 12%-48% for cerebral vascular health index; p < 0.05 for both), anti-dyslipidemic (13%-48% for skeletal muscle vascular health index; p < 0.05), anti-diabetic (12%-32% for cerebral vascular health index; p < 0.05) and anti-oxidant/inflammation (41%-64% for skeletal muscle vascular health index; 29%-42% for cerebral vascular health index; p < 0.05 for both) drugs. The results present the effectiveness of mechanistically diverse interventions to improve cerebral or skeletal muscle vascular health index in obese Zucker rats and provide insight into the superiority of some pharmacological agents despite similar effectiveness in terms of impact on intended targets. In addition, we demonstrate the utility of including a wider, more integrative approach to the study of microvasculopathy under settings of elevated disease risk and following pharmacological intervention. A major benefit of integrating vascular health index is an increased understanding of the development, timing and efficacy of interventions through greater insight into integrated microvascular function in combination with individual, higher resolution metrics.

Dual-center validation of using magnetic resonance imaging radiomics to predict stereotactic radiosurgery outcomes

Background

MRI radiomic features and machine learning have been used to predict brain metastasis (BM) stereotactic radiosurgery (SRS) outcomes. Previous studies used only single-center datasets, representing a significant barrier to clinical translation and further research. This study, therefore, presents the first dual-center validation of these techniques.

Methods

SRS datasets were acquired from 2 centers (n = 123 BMs and n = 117 BMs). Each dataset contained 8 clinical features, 107 pretreatment T1w contrast-enhanced MRI radiomic features, and post-SRS BM progression endpoints determined from follow-up MRI. Random decision forest models were used with clinical and/or radiomic features to predict progression. 250 bootstrap repetitions were used for single-center experiments.

Results

Training a model with one center's dataset and testing it with the other center's dataset required using a set of features important for outcome prediction at both centers, and achieved area under the receiver operating characteristic curve (AUC) values up to 0.70. A model training methodology developed using the first center's dataset was locked and externally validated with the second center's dataset, achieving a bootstrap-corrected AUC of 0.80. Lastly, models trained on pooled data from both centers offered balanced accuracy across centers with an overall bootstrap-corrected AUC of 0.78.

Conclusions

Using the presented validated methodology, radiomic models trained at a single center can be used externally, though they must utilize features important across all centers. These models' accuracies are inferior to those of models trained using each individual center's data. Pooling data across centers shows accurate and balanced performance, though further validation is required.

Using deep learning to predict tumor mutational burden from scans of H&E-stained multicenter slides of lung squamous cell carcinoma

Purpose

A high tumor mutational burden (TMB) is a promising biomarker for identifying lung squamous cell carcinoma (SqCC) patients who are more likely to benefit from risky but potentially highly beneficial immunotherapy, but it is not available in most clinics. It has been shown that it is possible to predict TMB from standard-of-care cancer histology slides using deep learning for various cancer sites. Our goal is to build a model that can do this specifically for lung SqCC and to validate it on a held-out test set from centers on which the model was not trained.

Approach

We obtained scans of diagnostic slides from 50 lung SqCC patients, with one slide per-patient, from 35 different centers. We held out 20 slides from 15 centers for testing and used the rest for training and validation, ensuring that no center was represented in more than one set. Using transfer learning, we explored several neural network architectures and training parameters to choose an optimal model.

Results

Using the training and validation sets, we found the optimal model to be VGG16. The per-patient AUC for this model on the held-out test set was 0.65, with an accuracy of 65%, true positive rate of 77%, and true negative rate of 43%.

Conclusions

A deep learning model can predict TMB from scans of H&E-stained slides of lung SqCC resections on an independent test set containing images only from centers on which the model was not trained. With further development and external validation, such a system can act as an alternative to traditional genetic sequencing for patients with SqCC; this will help physicians determine, with more accuracy, whether patients should be given immunotherapy. This will more effectively give access to immunotherapy drugs to those who need them and help spare others the toxicities associated with them.

A novel vascular health index: Using data analytics and population health to facilitate mechanistic modeling of microvascular status

The study of vascular function across conditions has been an intensive area of investigation for many years. While these efforts have revealed many factors contributing to vascular health, challenges remain for integrating results across research groups, animal models, and experimental conditions to understand integrated vascular function. As such, the insights attained in clinical/population research from linking datasets, have not been fully realized in the basic sciences, thus frustrating advanced analytics and complex modeling. To achieve comparable advances, we must address the conceptual challenge of defining/measuring integrated vascular function and the technical challenge of combining data across conditions, models, and groups. Here, we describe an approach to establish and validate a composite metric of vascular function by comparing parameters of vascular function in metabolic disease (the obese Zucker rat) to the same parameters in age-matched, "healthy" conditions, resulting in a common outcome measure which we term the vascular health index (VHI). VHI allows for the integration of datasets, thus expanding sample size and permitting advanced modeling to gain insight into the development of peripheral and cerebral vascular dysfunction. Markers of vascular reactivity, vascular wall mechanics, and microvascular network density are integrated in the VHI. We provide a detailed presentation of the development of the VHI and provide multiple measures to assess face, content, criterion, and discriminant validity of the metric. Our results demonstrate how the VHI captures multiple indices of dysfunction in the skeletal muscle and cerebral vasculature with metabolic disease and provide context for an integrated understanding of vascular health under challenged conditions.

Prostate cancer classification using radiomics and machine learning on mp-MRI validated using co-registered histology

BACKGROUND

Multi-parametric magnetic resonance imaging (mp-MRI) is emerging as a useful tool for prostate cancer (PCa) detection but currently has unaddressed limitations. Computer aided diagnosis (CAD) systems have been developed to address these needs, but many approaches used to generate and validate the models have inherent biases.

METHOD

All clinically significant PCa on histology was mapped to mp-MRI using a previously validated registration algorithm. Shape and size matched non-PCa regions were selected using a proposed sampling algorithm to eliminate biases towards shape and size. Further analysis was performed to assess biases regarding inter-zonal variability.

RESULTS

A 5-feature Naïve-Bayes classifier produced an area under the receiver operating characteristic curve (AUC) of 0.80 validated using leave-one-patient-out cross-validation. As mean inter-class area mismatch increased, median AUC trended towards positively biasing classifiers to producing higher AUCs. Classifiers were invariant to differences in shape between PCa and non-PCa lesions (AUC: 0.82 vs 0.82). Performance for models trained and tested only in the peripheral zone was found to be lower than in the central gland (AUC: 0.75 vs 0.95).

CONCLUSION

We developed a radiomics based machine learning system to classify PCa vs non-PCa tissue on mp-MRI validated on accurately co-registered mid-gland histology with a measured target registration error. Potential biases involved in model development were interrogated to provide considerations for future work in this area.

Arteriolar dysgenesis in ischemic, regenerating skeletal muscle revealed by automated micro-morphometry, computational modeling, and perfusion analysis

Rebuilding the local vasculature is central to restoring the health of muscles subjected to ischemic injury. Arteriogenesis yields remodeled collateral arteries that circumvent the obstruction, and angiogenesis produces capillaries to perfuse the regenerating myofibers. However, the vital intervening network of arterioles that feed the regenerated capillaries is poorly understood and an investigative challenge. We used machine learning and automated micro-morphometry to quantify the arteriolar landscape in distal hindlimb muscles in mice that have regenerated after femoral artery excision. Assessment of 1546 arteriolar sections revealed a striking (> 2-fold) increase in arteriolar density in regenerated muscle 14 and 28 days after ischemic injury. Lumen caliber was initially similar to that of control arterioles but after 4 weeks lumen area was reduced by 46%. In addition, the critical smooth muscle layer was attenuated throughout the arteriolar network, across a 150 to 5 µm diameter range. To understand the consequences of the reshaped distal hindlimb arterioles, we undertook computational flow modeling which revealed blunted flow augmentation. Moreover, impaired flow reserve was confirmed in vivo by laser Doppler analyses of flow in response to directly applied sodium nitroprusside. Thus, in hindlimb muscles regenerating after ischemic injury, the arteriolar network is amplified, inwardly remodels, and is diffusely under-muscularized. These defects and the associated flow restraints could contribute to the deleterious course of peripheral artery disease and merit attention when considering therapeutic innovations.

Prostate specific membrane antigen positron emission tomography for lesion-directed high-dose-rate brachytherapy dose escalation

•

This paper evaluated lesion-directed prostatic high dose rate brachytherapy.

•

Lesions defined by prostate specific membrane antigen positron emission tomography.

•

Dose escalation was confirmed using whole-mount digital histology.

•

Targeting lesions led to significantly higher dose to high-grade histologic cancer.

Background and purpose

Prostate specific membrane antigen positron emission tomography imaging (PSMA-PET) has demonstrated potential for intra-prostatic lesion localization. We leveraged our existing database of co-registered PSMA-PET imaging with cross sectional digitized pathology to model dose coverage of histologically-defined prostate cancer when tailoring brachytherapy dose escalation based on PSMA-PET imaging.

Materials and methods

Using a previously-developed automated approach, we created segmentation volumes delineating underlying dominant intraprostatic lesions for ten men with co-registered pathology-imaging datasets. To simulate realistic high-dose-rate brachytherapy (HDR-BT) treatments, we registered the PSMA-PET-defined segmentation volumes and underlying cancer to 3D trans-rectal ultrasound images of HDR-BT cases where 15 Gray (Gy) was delivered. We applied dose/volume optimization to focally target the dominant intraprostatic lesion identified on PSMA-PET. We then compared histopathology dose for all high-grade cancer within whole-gland treatment plans versus PSMA-PET-targeted plans. Histopathology dose was analyzed for all clinically significant cancer with a Gleason score of 7or greater.

Results

The standard whole-gland plans achieved a median [interquartile range] D98 of 15.2 [13.8–16.4] Gy to the histologically-defined cancer, while the targeted plans achieved a significantly higher D98 of 16.5 [15.0–19.0] Gy (p = 0.007).

Conclusion

This study is the first to use digital histology to confirm the effectiveness of PSMA-PET HDR-BT dose escalation using automatically generated contours. Based on the findings of this study, PSMA-PET lesion dose escalation can lead to increased dose to the ground truth histologically defined cancer.

A multiobserver study investigating the effectiveness of prostatic multiparametric magnetic resonance imaging to dose escalate corresponding histologic lesions using high-dose-rate brachytherapy

PURPOSE

Using multiparametric MRI data and the pathologic data from radical prostatectomy specimens, we simulated the treatment planning of dose-escalated high-dose-rate brachytherapy (HDR-BT) to the Multiparametric MRI dominant intraprostatic lesion (mpMRI-DIL) to compare the dose potentially delivered to the pathologically confirmed locations of the high-grade component of the cancer.

METHODS AND MATERIALS

Pathologist-annotated prostatectomy midgland histology sections from 12 patients were registered to preprostatectomy mpMRI scans that were interpreted by four radiologists. To simulate realistic HDR-BT, we registered each observer's mpMRI-DILs and corresponding histology to two transrectal ultrasound images of other HDR-BT patients with a 15-Gy whole-gland prescription. We used clinical inverse planning to escalate the mpMRI-DILs to 20.25 Gy. We compared the dose that the histopathology would have received if treated with standard treatment plans to the dose mpMRI-targeting would have achieved. The histopathology was grouped as high-grade cancer (any Gleason Grade 4 or 5) and low-grade cancer (only Gleason Grade 3).

RESULTS

212 mpMRI-targeted HDR-BT plans were analyzed. For high-grade histology, the mpMRI-targeted plans achieved significantly higher median [IQR] D98 and D90 values of 18.2 [16.7-19.5] Gy and 19.4 [17.8-20.9] Gy, respectively, in comparison with the standard plans (p = 0.01 and p = 0.003). For low-grade histology, the targeted treatment plans would have resulted in a significantly higher median D90 of 17.0 [16.1-18.4] Gy in comparison with standard plans (p = 0.015); the median D98 was not significantly higher (p = 0.2).

CONCLUSIONS

In this retrospective pilot study of 12 patients, mpMRI-based dose escalation led to increased dose to high-grade, but not low-grade, cancer. In our data set, different observers and mpMRI sequences had no substantial effect on dose to histologic cancer.

Histologic validation of auto-contoured dominant intraprostatic lesions on [18F] DCFPyL PSMA-PET imaging

BACKGROUND

PSMA-PET¹ has shown good concordance with histology, but there is a need to investigate the ability of PSMA-PET to delineate DIL² boundaries for guided biopsy and focal therapy planning.

OBJECTIVE

To determine threshold and margin combinations that satisfy the following criteria: ≥95% sensitivity with max specificity and ≥95% specificity with max sensitivity.

DESIGN, SETTING AND PARTICIPANTS

We registered pathologist-annotated whole-mount mid-gland prostatectomy histology sections cut in 4.4 mm intervals from 12 patients to pre-surgical PSMA-PET/MRI by mapping histology to ex-vivo imaging to in-vivo imaging. We generated PET-derived tumor volumes using boundaries defined by thresholded PET volumes from 1-100% of SUV³_max in 1% intervals. At each interval, we applied margins of 0-30 voxels in one voxel increments, giving 3000 volumes/patient.

OUTCOME MEASUREMENTS

Mean and standard deviation of sensitivity and specificity for cancer detection within the 2D oblique histologic planes that intersected with the 3D PET volume for each patient.

RESULTS AND LIMITATIONS

A threshold of 67% SUV max with an 8.4 mm margin achieved a (mean ± std.) sensitivity of 95.0 ± 7.8% and specificity of 76.4 ± 14.7%. A threshold of 81% SUV max with a 5.1 mm margin achieved sensitivity of 65.1 ± 28.4% and specificity of 95.1 ± 5.2%.

CONCLUSIONS

Preliminary evidence of thresholding and margin expansion of PSMA-PET images targeted at DILs validated with histopathology demonstrated excellent mean sensitivity and specificity in the setting of focal therapy/boosting and guided biopsy. These parameters can be used in a larger validation study supporting clinical translation.

Automatic cancer detection on digital histopathology images of mid-gland radical prostatectomy specimens

Purpose: Automatic cancer detection on radical prostatectomy (RP) sections facilitates graphical and quantitative surgical pathology reporting, which can potentially benefit postsurgery follow-up care and treatment planning. It can also support imaging validation studies using a histologic reference standard and pathology research studies. This problem is challenging due to the large sizes of digital histopathology whole-mount whole-slide images (WSIs) of RP sections and staining variability across different WSIs. Approach: We proposed a calibration-free adaptive thresholding algorithm, which compensates for staining variability and yields consistent tissue component maps (TCMs) of the nuclei, lumina, and other tissues. We used and compared three machine learning methods for classifying each cancer versus noncancer region of interest (ROI) throughout each WSI: (1) conventional machine learning methods and 14 texture features extracted from TCMs, (2) transfer learning with pretrained AlexNet fine-tuned by TCM ROIs, and (3) transfer learning with pretrained AlexNet fine-tuned with raw image ROIs. Results: The three methods yielded areas under the receiver operating characteristic curve of 0.96, 0.98, and 0.98, respectively, in leave-one-patient-out cross validation using 1.3 million ROIs from 286 mid-gland whole-mount WSIs from 68 patients. Conclusion: Transfer learning with the use of TCMs demonstrated state-of-the-art overall performance and is more stable with respect to sample size across different tissue types. For the tissue types involving Gleason 5 (most aggressive) cancer, it achieved the best performance compared to the other tested methods. This tool can be translated to clinical workflow to assist graphical and quantitative pathology reporting for surgical specimens upon further multicenter validation.

Measurement of the fluorescence lifetime of GFP in high refractive index levitated droplets using FLIM

Green fluorescent protein (GFP) is a widely used fluorescent probe in the life sciences and biosciences due to its high quantum yield and extinction coefficient, and its ability to bind to biological systems of interest. This study measures the fluorescence lifetime of GFP in sucrose/water solutions of known molarity in order to determine the refractive index dependent lifetime of GFP. A range of refractive indices from 1.43-1.53 were probed by levitating micron sized droplets composed of water/sucrose/GFP in an optical trap under well-constrained conditions of relative humidity. This setup allows for the first reported measurements of the fluorescence lifetime of GFP at refractive indices greater than 1.46. The results obtained at refractive indices less than 1.46 show good agreement with previous studies. Further experiments that trapped droplets of deionised water containing GFP allowed the hygroscopic properties of GFP to be measured. GFP is found to be mildly hygroscopic by mass, but the high ratio of molecular masses of GFP to water (ca. 1500 : 1) signifies that water uptake is large on a per-mole basis. Hygroscopic properties are verified using brightfield microscope imaging, of GFP droplets at low and high relative humidity, by measuring the humidity dependent droplet size. In addition, this experiment allowed the refractive index of pure GFP to be estimated for the first time (1.72 ± 0.07). This work provides reference data for future experiments involving GFP, especially for those conducted in high refractive index media. The work also demonstrates that GFP can be used as a probe for aerosol studies, which require determination of the refractive index of the aerosol of any shape.

Histologic tissue components provide major cues for machine learning-based prostate cancer detection and grading on prostatectomy specimens

Automatically detecting and grading cancerous regions on radical prostatectomy (RP) sections facilitates graphical and quantitative pathology reporting, potentially benefitting post-surgery prognosis, recurrence prediction, and treatment planning after RP. Promising results for detecting and grading prostate cancer on digital histopathology images have been reported using machine learning techniques. However, the importance and applicability of those methods have not been fully investigated. We computed three-class tissue component maps (TCMs) from the images, where each pixel was labeled as nuclei, lumina, or other. We applied seven different machine learning approaches: three non-deep learning classifiers with features extracted from TCMs, and four deep learning, using transfer learning with the 1) TCMs, 2) nuclei maps, 3) lumina maps, and 4) raw images for cancer detection and grading on whole-mount RP tissue sections. We performed leave-one-patient-out cross-validation against expert annotations using 286 whole-slide images from 68 patients. For both cancer detection and grading, transfer learning using TCMs performed best. Transfer learning using nuclei maps yielded slightly inferior overall performance, but the best performance for classifying higher-grade cancer. This suggests that 3-class TCMs provide the major cues for cancer detection and grading primarily using nucleus features, which are the most important information for identifying higher-grade cancer.

Ion Binding Properties of a Naturally Occurring Metalloantibody

LT1009 is a humanized version of murine LT1002 IgG1 that employs two bridging Ca²⁺ ions to bind its antigen, the biologically active lipid sphingosine-1-phosphate (S1P). We crystallized and determined the X-ray crystal structure of the LT1009 Fab fragment in 10 mM CaCl₂ and found that it binds two Ca²⁺ in a manner similar to its antigen-bound state. Flame atomic absorption spectroscopy (FAAS) confirmed that murine LT1002 also binds Ca²⁺ in solution and inductively-coupled plasma-mass spectrometry (ICP-MS) revealed that, although Ca²⁺ is preferred, LT1002 can bind Mg²⁺ and, to much lesser extent, Ba²⁺. Isothermal titration calorimetry (ITC) indicated that LT1002 binds two Ca²⁺ ions endothermically with a measured dissociation constant (K_D) of 171 μM. Protein and genome sequence analyses suggested that LT1002 is representative of a small class of confirmed and potential metalloantibodies and that Ca²⁺ binding is likely encoded for in germline variable chain genes. To test this hypothesis, we engineered, expressed, and purified a Fab fragment consisting of naïve murine germline-encoded light and heavy chain genes from which LT1002 is derived and observed that it binds Ca²⁺ in solution. We propose that LT1002 is representative of a class of naturally occurring metalloantibodies that are evolutionarily conserved across diverse mammalian genomes.

Chronic Obstructive Pulmonary Disease: Thoracic CT Texture Analysis and Machine Learning to Predict Pulmonary Ventilation

Background Fixed airflow limitation and ventilation heterogeneity are common in chronic obstructive pulmonary disease (COPD). Conventional noncontrast CT provides airway and parenchymal measurements but cannot be used to directly determine lung function. Purpose To develop, train, and test a CT texture analysis and machine-learning algorithm to predict lung ventilation heterogeneity in participants with COPD. Materials and Methods In this prospective study (ClinicalTrials.gov: NCT02723474; conducted from January 2010 to February 2017), participants were randomized to optimization (n = 1), training (n = 67), and testing (n = 27) data sets. Hyperpolarized (HP) helium 3 (³He) MRI ventilation maps were co-registered with thoracic CT to provide ground truth labels, and 87 quantitative imaging features were extracted and normalized to lung averages to generate 174 features. The volume-of-interest dimension and the training data sampling method were optimized to maximize the area under the receiver operating characteristic curve (AUC). Forward feature selection was performed to reduce the number of features; logistic regression, linear support vector machine, and quadratic support vector machine classifiers were trained through fivefold cross validation. The highest-performing classification model was applied to the test data set. Pearson coefficients were used to determine the relationships between the model, MRI, and pulmonary function measurements. Results The quadratic support vector machine performed best in training and was applied to the test data set. Model-predicted ventilation maps had an accuracy of 88% (95% confidence interval [CI]: 88%, 88%) and an AUC of 0.82 (95% CI: 0.82, 0.83) when the HP ³He MRI ventilation maps were used as the reference standard. Model-predicted ventilation defect percentage (VDP) was correlated with VDP at HP ³He MRI (r = 0.90, P < .001). Both model-predicted and HP ³He MRI VDP were correlated with forced expiratory volume in 1 second (FEV₁) (model: r = -0.65, P < .001; MRI: r = -0.70, P < .001), ratio of FEV₁ to forced vital capacity (model: r = -0.73, P < .001; MRI: r = -0.75, P < .001), diffusing capacity (model: r = -0.69, P < .001; MRI: r = -0.65, P < .001), and quality-of-life score (model: r = 0.59, P = .001; MRI: r = 0.65, P < .001). Conclusion Model-predicted ventilation maps generated by using CT textures and machine learning were correlated with MRI ventilation maps (r = 0.90, P < .001). © RSNA, 2019 Online supplemental material is available for this article. See also the editorial by Fain in this issue.

Accuracy Validation of an Automated Method for Prostate Segmentation in Magnetic Resonance Imaging

Three dimensional (3D) manual segmentation of the prostate on magnetic resonance imaging (MRI) is a laborious and time-consuming task that is subject to inter-observer variability. In this study, we developed a fully automatic segmentation algorithm for T2-weighted endorectal prostate MRI and evaluated its accuracy within different regions of interest using a set of complementary error metrics. Our dataset contained 42 T2-weighted endorectal MRI from prostate cancer patients. The prostate was manually segmented by one observer on all of the images and by two other observers on a subset of 10 images. The algorithm first coarsely localizes the prostate in the image using a template matching technique. Then, it defines the prostate surface using learned shape and appearance information from a set of training images. To evaluate the algorithm, we assessed the error metric values in the context of measured inter-observer variability and compared performance to that of our previously published semi-automatic approach. The automatic algorithm needed an average execution time of ∼60 s to segment the prostate in 3D. When compared to a single-observer reference standard, the automatic algorithm has an average mean absolute distance of 2.8 mm, Dice similarity coefficient of 82%, recall of 82%, precision of 84%, and volume difference of 0.5 cm³ in the mid-gland. Concordant with other studies, accuracy was highest in the mid-gland and lower in the apex and base. Loss of accuracy with respect to the semi-automatic algorithm was less than the measured inter-observer variability in manual segmentation for the same task.

A comparison of prostate tumor targeting strategies using magnetic resonance imaging-targeted, transrectal ultrasound-guided fusion biopsy

PURPOSE

Magnetic resonance imaging (MRI)-targeted, three-dimensional (3D) transrectal ultrasound (TRUS)-guided prostate biopsy aims to reduce the 21-47% false-negative rate of clinical two-dimensional (2D) TRUS-guided systematic biopsy, but continues to yield false-negative results. This may be improved via needle target optimization, accounting for guidance system errors and image registration errors. As an initial step toward the goal of optimized prostate biopsy targeting, we investigated how needle delivery error impacts tumor sampling probability for two targeting strategies.

METHODS

We obtained MRI and 3D TRUS images from 49 patients. A radiologist and radiology resident assessed these MR images and contoured 81 suspicious regions, yielding tumor surfaces that were registered to 3D TRUS. The biopsy system's root-mean-squared needle delivery error (RMSE) and systematic error were modeled using an isotropic 3D Gaussian distribution. We investigated two different prostate tumor-targeting strategies using (a) the tumor's centroid and (b) a ring in the lateral-elevational plane. For each simulation, targets were spaced at equal arc lengths on a ring with radius equal to the systematic error magnitude. A total of 1000 biopsy simulations were conducted for each tumor, with RMSE and systematic error magnitudes ranging from 1 to 6 mm. The difference in median tumor sampling probability and probability of obtaining a 50% core involvement was determined for ring vs centroid targeting.

RESULTS

Our simulation results indicate that ring targeting outperformed centroid targeting in situations where systematic error exceeds RMSE. In these instances, we observed statistically significant differences showing 1-32% improvement in sampling probability due to ring targeting. Likewise, we observed statistically significant differences showing 1-39% improvement in 50% core involvement probability due to ring targeting.

CONCLUSIONS

Our results suggest that the optimal targeting scheme for prostate biopsy depends on the relative levels of systematic and random errors in the system. Where systematic error dominates, a ring-targeting scheme may yield improved probability of tumor sampling. The findings presented in this paper may be used to aid in target selection strategies for clinicians performing targeted prostate biopsies on any MRI targeted, 3D TRUS-guided biopsy system and could support earlier diagnosis of prostate cancer while it remains localized to the gland and curable.

Robust 2-D-3-D Registration Optimization for Motion Compensation During 3-D TRUS-Guided Biopsy Using Learned Prostate Motion Data

In magnetic resonance (MR)-targeted, 3-D transrectal ultrasound (TRUS)-guided biopsy, prostate motion during the procedure increases the needle targeting error and limits the ability to accurately sample MR-suspicious tumor volumes. The robustness of the 2-D-3-D registration methods for prostate motion compensation is impacted by local optima in the search space. In this paper, we analyzed the prostate motion characteristics and investigated methods to incorporate such knowledge into the registration optimization framework to improve robustness against local optima. Rigid motion of the prostate was analyzed adopting a mixture-of-Gaussian (MoG) model using 3-D TRUS images acquired at bilateral sextant probe positions with a mechanically assisted biopsy system. The learned motion characteristics were incorporated into Powell's direction set method by devising multiple initial search positions and initial search directions. Experiments were performed on data sets acquired during clinical biopsy procedures, and registration error was evaluated using target registration error (TRE) and converged image similarity metric values after optimization. After incorporating the learned initialization positions and directions in Powell's method, 2-D-3-D registration to compensate for motion during prostate biopsy was performed with rms ± std TRE of 2.33 ± 1.09 mm with ~3 s mean execution time per registration. This was an improvement over 3.12 ± 1.70 mm observed in Powell's standard approach. For the data acquired under clinical protocols, the converged image similarity metric value improved in ≥8% of the registrations whereas it degraded only ≤1% of the registrations. The reported improvements in optimization indicate useful advancements in robustness to ensure smooth clinical integration of a registration solution for motion compensation that facilitates accurate sampling of the smallest clinically significant tumors.

Differentiation of arterioles from venules in mouse histology images using machine learning

Analysis and morphological comparison of the arteriolar and venular components of a microvascular network are essential to our understanding of multiple diseases affecting every organ system. We have developed and evaluated the first fully automatic software system for differentiation of arterioles from venules on high-resolution digital histology images of the mouse hind limb immunostained with smooth muscle [Formula: see text]-actin. Classifiers trained on statistical and morphological features by supervised machine learning provided useful classification accuracy for differentiation of arterioles from venules, achieving an area under the receiver operating characteristic curve of 0.89. Feature selection was consistent across cross validation iterations, and a small set of two features was required to achieve the reported performance, suggesting the generalizability of the system. This system eliminates the need for laborious manual classification of the hundreds of microvessels occurring in a typical sample and paves the way for high-throughput analysis of the arteriolar and venular networks in the mouse.

Segmentation of digitized histological sections for quantification of the muscularized vasculature in the mouse hind limb

Immunohistochemical tissue staining enhances microvasculature characteristics, including the smooth muscle in the medial layer of the vessel walls that is responsible for regulation of blood flow. The vasculature can be imaged in a comprehensive fashion using whole-slide scanning. However, since each such image potentially contains hundreds of small vessels, manual vessel delineation and quantification is not practically feasible. In this work, we present a fully automatic segmentation and vasculature quantification algorithm for whole-slide images. We evaluated its performance on tissue samples drawn from the hind limbs of wild-type mice, stained for smooth muscle using 3,3'-Diaminobenzidine (DAB) immunostain. The algorithm was designed to be robust to vessel fragmentation due to staining irregularity, and artefactual staining of nonvessel objects. Colour deconvolution was used to isolate the DAB stain for detection of vessel wall fragments. Complete vessels were reconstructed from the fragments by joining endpoints of topological skeletons. Automatic measures of vessel density, perimeter, wall area and local wall thickness were taken. The segmentation algorithm was validated against manual measures, resulting in a Dice similarity coefficient of 89%. The relationships observed between these measures were as expected from a biological standpoint, providing further reinforcement of the accuracy of this system. This system provides a fully automated and accurate means of measuring the arteriolar and venular morphology of vascular smooth muscle.

Measurement of the Raman spectra and hygroscopicity of four pharmaceutical aerosols as they travel from pressurised metered dose inhalers (pMDI) to a model lung

Particle inhalation is an effective and rapid delivery method for a variety of pharmaceuticals, particularly bronchodilation drugs used for treating asthma and COPD. Conditions of relative humidity and temperature inside the lungs are generally very different from the outside ambient air, with the lung typically being warmer and more humid. Changes in humidity, from inhaler to lung, can cause hygroscopic phase transitions and particle growth. Increasing particle size and mass can negatively affect particle deposition within the lung leading to inefficient treatment, while deliquescence prior to impaction is liable to accelerate drug uptake. To better understand the hygroscopic properties of four pharmaceutical aerosol particles; pharmaceutical particles from four commercially available pressurised metered dose inhalers (pMDIs) were stably captured in an optical trap, and their composition was examined online via Raman spectroscopy. Micron-sized particles of salbutamol sulfate, salmeterol xinafoate, fluticasone propionate and ciclesonide were levitated and examined over a range of relative humidity values inside a chamber designed to mimic conditions within the respiratory tract. The effect of temperature upon hygroscopicity was also investigated for salbutamol sulfate particles. Salbutamol sulfate was found to have significant hygroscopicity, salmeterol xinafoate showed some hygroscopic interactions, whilst fluticasone propionate and ciclesonide revealed no observable hygroscopicity. Thermodynamic and structural modelling is used to explain the observed experimental results.

Postediting prostate magnetic resonance imaging segmentation consistency and operator time using manual and computer-assisted segmentation: multiobserver study

Prostate segmentation on T2w MRI is important for several diagnostic and therapeutic procedures for prostate cancer. Manual segmentation is time-consuming, labor-intensive, and subject to high interobserver variability. This study investigated the suitability of computer-assisted segmentation algorithms for clinical translation, based on measurements of interoperator variability and measurements of the editing time required to yield clinically acceptable segmentations. A multioperator pilot study was performed under three pre- and postediting conditions: manual, semiautomatic, and automatic segmentation. We recorded the required editing time for each segmentation and measured the editing magnitude based on five different spatial metrics. We recorded average editing times of 213, 328, and 393 s for manual, semiautomatic, and automatic segmentation respectively, while an average fully manual segmentation time of 564 s was recorded. The reduced measured postediting interoperator variability of semiautomatic and automatic segmentations compared to the manual approach indicates the potential of computer-assisted segmentation for generating a clinically acceptable segmentation faster with higher consistency. The lack of strong correlation between editing time and the values of typically used error metrics ([Formula: see text]) implies that the necessary postsegmentation editing time needs to be measured directly in order to evaluate an algorithm's suitability for clinical translation.

Comparison of prostate MRI-3D transrectal ultrasound fusion biopsy for first-time and repeat biopsy patients with previous atypical small acinar proliferation

INTRODUCTION

This study evaluates the clinical benefit of magnetic resonance-transrectal ultrasound (MR-TRUS) fusion biopsy over systematic biopsy between first-time and repeat prostate biopsy patients with prior atypical small acinar proliferation (ASAP).

MATERIALS

100 patients were enrolled in a single-centre prospective cohort study: 50 for first biopsy, 50 for repeat biopsy with prior ASAP. Multiparameteric magnetic resonance imaging (MP-MRI) and standard 12-core ultrasound biopsy (Std-Bx) were performed on all patients. Targeted biopsy using MRI-TRUS fusion (Fn-Bx) was performed f suspicious lesions were identified on the pre-biopsy MP-MRI. Classification of clinically significant disease was assessed independently for the Std-Bx vs. Fn-Bx cores to compare the two approaches.

RESULTS

Adenocarcinoma was detected in 49/100 patients (26 first biopsy, 23 ASAP biopsy), with 25 having significant disease (17 first, 8 ASAP). Fn-Bx demonstrated significantly higher per-core cancer detection rates, cancer involvement, and Gleason scores for first-time and ASAP patients. However, Fn-Bx was significantly more likely to detect significant cancer missed on Std-Bx for ASAP patients than first-time biopsy patients. The addition of Fn-Bx to Std-Bx for ASAP patients had a 166.7% relative risk reduction for missing Gleason ≥ 3 + 4 disease (number needed to image with MP-MRI=10 patients) compared to 6.3% for first biopsy (number to image=50 patients). Negative predictive value of MP-MRI for negative biopsy was 79% for first-time and 100% for ASAP patients, with median followup of 32.1 ± 15.5 months.

CONCLUSIONS

MR-TRUS Fn-Bx has a greater clinical impact for repeat biopsy patients with prior ASAP than biopsy-naïve patients by detecting more significant cancers that are missed on Std-Bx.

Pulmonary imaging after stereotactic radiotherapy-does RECIST still apply?

The use of stereotactic ablative radiotherapy (SABR) for the treatment of primary lung cancer and metastatic disease is rapidly increasing. However, the presence of benign fibrotic changes on CT imaging makes response assessment following SABR a challenge, as these changes develop with an appearance similar to tumour recurrence. Misclassification of benign fibrosis as local recurrence has resulted in unnecessary interventions, including biopsy and surgical resection. Response evaluation criteria in solid tumours (RECIST) are widely used as a universal set of guidelines to assess tumour response following treatment. However, in the context of non-spherical and irregular post-SABR fibrotic changes, the RECIST criteria can have several limitations. Positron emission tomography can also play a role in response assessment following SABR; however, false-positive results in regions of inflammatory lung post-SABR can be a major clinical issue and optimal standardized uptake values to distinguish fibrosis and recurrence have not been determined. Although validated CT high-risk features show a high sensitivity and specificity for predicting recurrence, most recurrences are not detected until more than 1-year post-treatment. Advanced quantitative radiomic analysis on CT imaging has demonstrated promise in distinguishing benign fibrotic changes from local recurrence at earlier time points, and more accurately, than physician assessment. Overall, the use of RECIST alone may prove inferior to novel metrics of assessing response.

Toward Prostate Cancer Contouring Guidelines on Magnetic Resonance Imaging: Dominant Lesion Gross and Clinical Target Volume Coverage Via Accurate Histology Fusion

PURPOSE

Defining prostate cancer (PCa) lesion clinical target volumes (CTVs) for multiparametric magnetic resonance imaging (mpMRI) could support focal boosting or treatment to improve outcomes or lower morbidity, necessitating appropriate CTV margins for mpMRI-defined gross tumor volumes (GTVs). This study aimed to identify CTV margins yielding 95% coverage of PCa tumors for prospective cases with high likelihood.

METHODS AND MATERIALS

Twenty-five men with biopsy-confirmed clinical stage T1 or T2 PCa underwent pre-prostatectomy mpMRI, yielding T2-weighted, dynamic contrast-enhanced, and apparent diffusion coefficient images. Digitized whole-mount histology was contoured and registered to mpMRI scans (error ≤2 mm). Four observers contoured lesion GTVs on each mpMRI scan. CTVs were defined by isotropic and anisotropic expansion from these GTVs and from multiparametric (unioned) GTVs from 2 to 3 scans. Histologic coverage (proportions of tumor area on co-registered histology inside the CTV, measured for Gleason scores [GSs] ≥6 and ≥7) and prostate sparing (proportions of prostate volume outside the CTV) were measured. Nonparametric histologic-coverage prediction intervals defined minimal margins yielding 95% coverage for prospective cases with 78% to 92% likelihood.

RESULTS

On analysis of 72 true-positive tumor detections, 95% coverage margins were 9 to 11 mm (GS ≥ 6) and 8 to 10 mm (GS ≥ 7) for single-sequence GTVs and were 8 mm (GS ≥ 6) and 6 mm (GS ≥ 7) for 3-sequence GTVs, yielding CTVs that spared 47% to 81% of prostate tissue for the majority of tumors. Inclusion of T2-weighted contours increased sparing for multiparametric CTVs with 95% coverage margins for GS ≥6, and inclusion of dynamic contrast-enhanced contours increased sparing for GS ≥7. Anisotropic 95% coverage margins increased the sparing proportions to 71% to 86%.

CONCLUSIONS

Multiparametric magnetic resonance imaging-defined GTVs expanded by appropriate margins may support focal boosting or treatment of PCa; however, these margins, accounting for interobserver and intertumoral variability, may preclude highly conformal CTVs. Multiparametric GTVs and anisotropic margins may reduce the required margins and improve prostate sparing.

Imaging texture analysis for automated prediction of lung cancer recurrence after stereotactic radiotherapy

Benign radiation-induced lung injury (RILI) is not uncommon following stereotactic ablative radiotherapy (SABR) for lung cancer and can be difficult to differentiate from tumor recurrence on follow-up imaging. We previously showed the ability of computed tomography (CT) texture analysis to predict recurrence. The aim of this study was to evaluate and compare the accuracy of recurrence prediction using manual region-of-interest segmentation to that of a semiautomatic approach. We analyzed 22 patients treated for 24 lesions (11 recurrences, 13 RILI). Consolidative and ground-glass opacity (GGO) regions were manually delineated. The longest axial diameter of the consolidative region on each post-SABR CT image was measured. This line segment is routinely obtained as part of the clinical imaging workflow and was used as input to automatically delineate the consolidative region and subsequently derive a periconsolidative region to sample GGO tissue. Texture features were calculated, and at two to five months post-SABR, the entropy texture measure within the semiautomatic segmentations showed prediction accuracies [areas under the receiver operating characteristic curve (AUC): 0.70 to 0.73] similar to those of manual GGO segmentations (AUC: 0.64). After integration into the clinical workflow, this decision support system has the potential to support earlier salvage for patients with recurrence and fewer investigations of benign RILI.

Computer-Aided Prostate Cancer Detection Using Ultrasound RF Time Series: In Vivo Feasibility Study

This paper presents the results of a computer-aided intervention solution to demonstrate the application of RF time series for characterization of prostate cancer, in vivo.

METHODS

We pre-process RF time series features extracted from 14 patients using hierarchical clustering to remove possible outliers. Then, we demonstrate that the mean central frequency and wavelet features extracted from a group of patients can be used to build a nonlinear classifier which can be applied successfully to differentiate between cancerous and normal tissue regions of an unseen patient.

RESULTS

In a cross-validation strategy, we show an average area under receiver operating characteristic curve (AUC) of 0.93 and classification accuracy of 80%. To validate our results, we present a detailed ultrasound to histology registration framework.

CONCLUSION

Ultrasound RF time series results in differentiation of cancerous and normal tissue with high AUC.

Fibroblast Growth Factor 9 Imparts Hierarchy and Vasoreactivity to the Microcirculation of Renal Tumors and Suppresses Metastases

Tumor vessel normalization has been proposed as a therapeutic paradigm. However, normal microvessels are hierarchical and vasoreactive with single file transit of red blood cells through capillaries. Such a network has not been identified in malignant tumors. We tested whether the chaotic tumor microcirculation could be reconfigured by the mesenchyme-selective growth factor, FGF9. Delivery of FGF9 to renal tumors in mice yielded microvessels that were covered by pericytes, smooth muscle cells, and a collagen-fortified basement membrane. This was associated with reduced pulmonary metastases. Intravital microvascular imaging revealed a haphazard web of channels in control tumors but a network of arterioles, bona fide capillaries, and venules in FGF9-expressing tumors. Moreover, whereas vasoreactivity was absent in control tumors, arterioles in FGF9-expressing tumors could constrict and dilate in response to adrenergic and nitric oxide releasing agents, respectively. These changes were accompanied by reduced hypoxia in the tumor core and reduced expression of the angiogenic factor VEGF-A. FGF9 was found to selectively amplify a population of PDGFRβ-positive stromal cells in the tumor and blocking PDGFRβ prevented microvascular differentiation by FGF9 and also worsened metastases. We conclude that harnessing local mesenchymal stromal cells with FGF9 can differentiate the tumor microvasculature to an extent not observed previously.

Spatially varying accuracy and reproducibility of prostate segmentation in magnetic resonance images using manual and semiautomated methods

PURPOSE

Three-dimensional (3D) prostate image segmentation is useful for cancer diagnosis and therapy guidance, but can be time-consuming to perform manually and involves varying levels of difficulty and interoperator variability within the prostatic base, midgland (MG), and apex. In this study, the authors measured accuracy and interobserver variability in the segmentation of the prostate on T2-weighted endorectal magnetic resonance (MR) imaging within the whole gland (WG), and separately within the apex, midgland, and base regions.

METHODS

The authors collected MR images from 42 prostate cancer patients. Prostate border delineation was performed manually by one observer on all images and by two other observers on a subset of ten images. The authors used complementary boundary-, region-, and volume-based metrics [mean absolute distance (MAD), Dice similarity coefficient (DSC), recall rate, precision rate, and volume difference (ΔV)] to elucidate the different types of segmentation errors that they observed. Evaluation for expert manual and semiautomatic segmentation approaches was carried out. Compared to manual segmentation, the authors' semiautomatic approach reduces the necessary user interaction by only requiring an indication of the anteroposterior orientation of the prostate and the selection of prostate center points on the apex, base, and midgland slices. Based on these inputs, the algorithm identifies candidate prostate boundary points using learned boundary appearance characteristics and performs regularization based on learned prostate shape information.

RESULTS

The semiautomated algorithm required an average of 30 s of user interaction time (measured for nine operators) for each 3D prostate segmentation. The authors compared the segmentations from this method to manual segmentations in a single-operator (mean whole gland MAD = 2.0 mm, DSC = 82%, recall = 77%, precision = 88%, and ΔV = - 4.6 cm(3)) and multioperator study (mean whole gland MAD = 2.2 mm, DSC = 77%, recall = 72%, precision = 86%, and ΔV = - 4.0 cm(3)). These results compared favorably with observed differences between manual segmentations and a simultaneous truth and performance level estimation reference for this data set (whole gland differences as high as MAD = 3.1 mm, DSC = 78%, recall = 66%, precision = 77%, and ΔV = 15.5 cm(3)). The authors found that overall, midgland segmentation was more accurate and repeatable than the segmentation of the apex and base, with the base posing the greatest challenge.

CONCLUSIONS

The main conclusions of this study were that (1) the semiautomated approach reduced interobserver segmentation variability; (2) the segmentation accuracy of the semiautomated approach, as well as the accuracies of recently published methods from other groups, were within the range of observed expert variability in manual prostate segmentation; and (3) further efforts in the development of computer-assisted segmentation would be most productive if focused on improvement of segmentation accuracy and reduction of variability within the prostatic apex and base.

A Method for 3D Histopathology Reconstruction Supporting Mouse Microvasculature Analysis

Structural abnormalities of the microvasculature can impair perfusion and function. Conventional histology provides good spatial resolution with which to evaluate the microvascular structure but affords no 3-dimensional information; this limitation could lead to misinterpretations of the complex microvessel network in health and disease. The objective of this study was to develop and evaluate an accurate, fully automated 3D histology reconstruction method to visualize the arterioles and venules within the mouse hind-limb. Sections of the tibialis anterior muscle from C57BL/J6 mice (both normal and subjected to femoral artery excision) were reconstructed using pairwise rigid and affine registrations of 5 µm-thick, paraffin-embedded serial sections digitized at 0.25 µm/pixel. Low-resolution intensity-based rigid registration was used to initialize the nucleus landmark-based registration, and conventional high-resolution intensity-based registration method. The affine nucleus landmark-based registration was developed in this work and was compared to the conventional affine high-resolution intensity-based registration method. Target registration errors were measured between adjacent tissue sections (pairwise error), as well as with respect to a 3D reference reconstruction (accumulated error, to capture propagation of error through the stack of sections). Accumulated error measures were lower (p < 0.01) for the nucleus landmark technique and superior vasculature continuity was observed. These findings indicate that registration based on automatic extraction and correspondence of small, homologous landmarks may support accurate 3D histology reconstruction. This technique avoids the otherwise problematic "banana-into-cylinder" effect observed using conventional methods that optimize the pairwise alignment of salient structures, forcing them to be section-orthogonal. This approach will provide a valuable tool for high-accuracy 3D histology tissue reconstructions for analysis of diseased microvasculature.

Abstract 431: Conversion of Tumor Microvessels into a Hierarchical and Vasoreactive Network, and Suppression of Metastases, by Fibroblast Growth Factor 9

Normalizing the tumor vasculature has been proposed as a therapeutic paradigm. However, to function normally, microvessels must exist as a vasoreactive and hierarchical network with red blood cells flowing single file through capillaries. Such a network has not been identified in malignant tumors. We previously found that fibroblast growth factor 9 (FGF9) could stabilize new blood vessels in ischemic muscle. To determine if FGF9 impacted tumors vessels, renal carcinoma (Renca) cells, expressing GFP or FGF9, were implanted into the subcapsular space of female Balb/c mice. After 14 days, the resulting FGF9-tumors had 17% fewer microvessels than control tumors (p=0.003) but the vessels had a collagen-fortified basement membrane and were more extensively covered with pericytes (4-fold, p=0.015) and smooth muscle cells (14-fold, p=0.002). Notably, this was associated with reduced pulmonary metastases (p=0.029). Intravital video microscopy revealed that FGF9 converted a haphazard web of channels into a hierarchal network with arterioles, capillaries, and venules. There was also a 33% reduction in vessel length density (p=0.034), a 67% reduction in mean lumen diameter (p<0.001), and 57% fewer bifurcations (p=0.019). Moreover, whereas vasoreactivity was absent in control tumors, arterioles in FGF9-tumors could constrict and dilate in response to adrenergic and nitric oxide releasing agents, respectively. Pimonidazole infusion revealed a 33% reduction of hypoxia in the tumor core (p=0.031) with a 35% reduction in VEGFA expression (p=0.031). Immunostaining and selective cell harvesting revealed that FGF9 selectively amplified a population of PDGFRß-positive stromal cells in the tumor (p=0.045). Furthermore, in vivo blocking of PDGFRß prevented microvascular differentiation by FGF9 and worsened metastases (p=0.002).

Conclusion: FGF9 can impart an otherwise dysfunctional tumor microvasculature with hierarchy, vasoreactivity, and improved oxygen delivery, via selective amplification of PDGFRß-expressing mesenchymal stromal cells. These findings suggest an approach to driving microvascular network differentiation, to an extent not observed previously, to pacify the tumor.

Magnetic resonance imaging-targeted, 3D transrectal ultrasound-guided fusion biopsy for prostate cancer: Quantifying the impact of needle delivery error on diagnosis

PURPOSE

Magnetic resonance imaging (MRI)-targeted, 3D transrectal ultrasound (TRUS)-guided "fusion" prostate biopsy intends to reduce the ∼23% false negative rate of clinical two-dimensional TRUS-guided sextant biopsy. Although it has been reported to double the positive yield, MRI-targeted biopsies continue to yield false negatives. Therefore, the authors propose to investigate how biopsy system needle delivery error affects the probability of sampling each tumor, by accounting for uncertainties due to guidance system error, image registration error, and irregular tumor shapes.

METHODS

T2-weighted, dynamic contrast-enhanced T1-weighted, and diffusion-weighted prostate MRI and 3D TRUS images were obtained from 49 patients. A radiologist and radiology resident contoured 81 suspicious regions, yielding 3D tumor surfaces that were registered to the 3D TRUS images using an iterative closest point prostate surface-based method to yield 3D binary images of the suspicious regions in the TRUS context. The probabilityP of obtaining a sample of tumor tissue in one biopsy core was calculated by integrating a 3D Gaussian distribution over each suspicious region domain. Next, the authors performed an exhaustive search to determine the maximum root mean squared error (RMSE, in mm) of a biopsy system that gives P ≥ 95% for each tumor sample, and then repeated this procedure for equal-volume spheres corresponding to each tumor sample. Finally, the authors investigated the effect of probe-axis-direction error on measured tumor burden by studying the relationship between the error and estimated percentage of core involvement.

RESULTS

Given a 3.5 mm RMSE for contemporary fusion biopsy systems,P ≥ 95% for 21 out of 81 tumors. The authors determined that for a biopsy system with 3.5 mm RMSE, one cannot expect to sample tumors of approximately 1 cm(3) or smaller with 95% probability with only one biopsy core. The predicted maximum RMSE giving P ≥ 95% for each tumor was consistently greater when using spherical tumor shapes as opposed to no shape assumption. However, an assumption of spherical tumor shape for RMSE = 3.5 mm led to a mean overestimation of tumor sampling probabilities of 3%, implying that assuming spherical tumor shape may be reasonable for many prostate tumors. The authors also determined that a biopsy system would need to have a RMS needle delivery error of no more than 1.6 mm in order to sample 95% of tumors with one core. The authors' experiments also indicated that the effect of axial-direction error on the measured tumor burden was mitigated by the 18 mm core length at 3.5 mm RMSE.

CONCLUSIONS

For biopsy systems with RMSE ≥ 3.5 mm, more than one biopsy core must be taken from the majority of tumors to achieveP ≥ 95%. These observations support the authors' perspective that some tumors of clinically significant sizes may require more than one biopsy attempt in order to be sampled during the first biopsy session. This motivates the authors' ongoing development of an approach to optimize biopsy plans with the aim of achieving a desired probability of obtaining a sample from each tumor, while minimizing the number of biopsies. Optimized planning of within-tumor targets for MRI-3D TRUS fusion biopsy could support earlier diagnosis of prostate cancer while it remains localized to the gland and curable.

Ultrasound-Based Characterization of Prostate Cancer Using Joint Independent Component Analysis

OBJECTIVE

This paper presents the results of a new approach for selection of RF time series features based on joint independent component analysis for in vivo characterization of prostate cancer.

METHODS

We project three sets of RF time series features extracted from the spectrum, fractal dimension, and the wavelet transform of the ultrasound RF data on a space spanned by five joint independent components. Then, we demonstrate that the obtained mixing coefficients from a group of patients can be used to train a classifier, which can be applied to characterize cancerous regions of a test patient.

RESULTS

In a leave-one-patient-out cross validation, an area under receiver operating characteristic curve of 0.93 and classification accuracy of 84% are achieved.

CONCLUSION

Ultrasound RF time series can be used to accurately characterize prostate cancer, in vivo without the need for exhaustive search in the feature space.

SIGNIFICANCE

We use joint independent component analysis for systematic fusion of multiple sets of RF time series features, within a machine learning framework, to characterize PCa in an in vivo study.

Early prediction of tumor recurrence based on CT texture changes after stereotactic ablative radiotherapy (SABR) for lung cancer

PURPOSE

Benign computed tomography (CT) changes due to radiation induced lung injury (RILI) are common following stereotactic ablative radiotherapy (SABR) and can be difficult to differentiate from tumor recurrence. The authors measured the ability of CT image texture analysis, compared to more traditional measures of response, to predict eventual cancer recurrence based on CT images acquired within 5 months of treatment.

METHODS

A total of 24 lesions from 22 patients treated with SABR were selected for this study: 13 with moderate to severe benign RILI, and 11 with recurrence. Three-dimensional (3D) consolidative and ground-glass opacity (GGO) changes were manually delineated on all follow-up CT scans. Two size measures of the consolidation regions (longest axial diameter and 3D volume) and nine appearance features of the GGO were calculated: 2 first-order features [mean density and standard deviation of density (first-order texture)], and 7 second-order texture features [energy, entropy, correlation, inverse difference moment (IDM), inertia, cluster shade, and cluster prominence]. For comparison, the corresponding response evaluation criteria in solid tumors measures were also taken for the consolidation regions. Prediction accuracy was determined using the area under the receiver operating characteristic curve (AUC) and two-fold cross validation (CV).

RESULTS

For this analysis, 46 diagnostic CT scans scheduled for approximately 3 and 6 months post-treatment were binned based on their recorded scan dates into 2-5 month and 5-8 month follow-up time ranges. At 2-5 months post-treatment, first-order texture, energy, and entropy provided AUCs of 0.79-0.81 using a linear classifier. On two-fold CV, first-order texture yielded 73% accuracy versus 76%-77% with the second-order features. The size measures of the consolidative region, longest axial diameter and 3D volume, gave two-fold CV accuracies of 60% and 57%, and AUCs of 0.72 and 0.65, respectively.

CONCLUSIONS

Texture measures of the GGO appearance following SABR demonstrated the ability to predict recurrence in individual patients within 5 months of SABR treatment. Appearance changes were also shown to be more accurately predictive of recurrence, as compared to size measures within the same time period. With further validation, these results could form the substrate for a clinically useful computer-aided diagnosis tool which could provide earlier salvage of patients with recurrence.

New techniques for assessing response after hypofractionated radiotherapy for lung cancer

Hypofractionated radiotherapy (HFRT) is an effective and increasingly-used treatment for early stage non-small cell lung cancer (NSCLC). Stereotactic ablative radiotherapy (SABR) is a form of HFRT and delivers biologically effective doses (BEDs) in excess of 100 Gy10 in 3-8 fractions. Excellent long-term outcomes have been reported; however, response assessment following SABR is complicated as radiation induced lung injury can appear similar to a recurring tumor on CT. Current approaches to scoring treatment responses include Response Evaluation Criteria in Solid Tumors (RECIST) and positron emission tomography (PET), both of which appear to have a limited role in detecting recurrences following SABR. Novel approaches to assess response are required, but new techniques should be easily standardized across centers, cost effective, with sensitivity and specificity that improves on current CT and PET approaches. This review examines potential novel approaches, focusing on the emerging field of quantitative image feature analysis, to distinguish recurrence from fibrosis after SABR.

3D prostate histology reconstruction: an evaluation of image-based and fiducial-based algorithms

PURPOSE

Evaluation of in vivo prostate imaging modalities for determining the spatial distribution and aggressiveness of prostate cancer ideally requires accurate registration of images to an accepted reference standard, such as histopathological examination of radical prostatectomy specimens. Three-dimensional (3D) reconstruction of prostate histology facilitates these registration-based evaluations by reintroducing 3D spatial information lost during histology processing. Because the reconstruction accuracy may constrain the clinical questions that can be answered with these data, it is important to assess the tradeoffs between minimally disruptive methods based on intrinsic image information and potentially more robust methods based on extrinsic fiducial markers.

METHODS

Ex vivo magnetic resonance (MR) images and digitized whole-mount histology images from 12 radical prostatectomy specimens were used to evaluate four 3D histology reconstruction algorithms. 3D reconstructions were computed by registering each histology image to the corresponding ex vivo MR image using one of two similarity metrics (mutual information or fiducial registration error) and one of two search domains (affine transformations or a constrained subset thereof). The algorithms were evaluated for accuracy using the mean target registration error (TRE) computed from homologous intrinsic point landmarks (3-16 per histology section; 232 total) identified on histology and MR images, and for the sensitivity of TRE to rotational, translational, and scaling initialization errors.

RESULTS

The algorithms using fiducial registration error and mutual information had mean ± standard deviation TREs of 0.7 ± 0.4 and 1.2 ± 0.7 mm, respectively, and one algorithm using fiducial registration error and affine transforms had negligible sensitivities to initialization errors. The postoptimization values of the mutual information-based metric showed evidence of errors due to both the optimizer and the similarity metric, and variation of parameters of the mutual information-based metric did not improve its performance.

CONCLUSIONS

The extrinsic fiducial-based algorithm had lower mean TRE and lower sensitivity to initialization than the intrinsic intensity-based algorithm using mutual information. A model relating statistical power to registration error for certain imaging validation study designs estimated that a reconstruction algorithm with a mean TRE of 0.7 mm would require 27% fewer subjects than the method used to initialize the algorithms (mean TRE 1.3 ± 0.7 mm), suggesting the choice of reconstruction technique can have a substantial impact on the design of imaging validation studies, and on their overall cost.

3D prostate histology image reconstruction: Quantifying the impact of tissue deformation and histology section location

Background:

Guidelines for localizing prostate cancer on imaging are ideally informed by registered post-prostatectomy histology. 3D histology reconstruction methods can support this by reintroducing 3D spatial information lost during histology processing. The need to register small, high-grade foci drives a need for high accuracy. Accurate 3D reconstruction method design is impacted by the answers to the following central questions of this work. (1) How does prostate tissue deform during histology processing? (2) What spatial misalignment of the tissue sections is induced by microtome cutting? (3) How does the choice of reconstruction model affect histology reconstruction accuracy?

Materials and Methods:

Histology, paraffin block face and magnetic resonance images were acquired for 18 whole mid-gland tissue slices from six prostates. 7-15 homologous landmarks were identified on each image. Tissue deformation due to histology processing was characterized using the target registration error (TRE) after landmark-based registration under four deformation models (rigid, similarity, affine and thin-plate-spline [TPS]). The misalignment of histology sections from the front faces of tissue slices was quantified using manually identified landmarks. The impact of reconstruction models on the TRE after landmark-based reconstruction was measured under eight reconstruction models comprising one of four deformation models with and without constraining histology images to the tissue slice front faces.

Results:

Isotropic scaling improved the mean TRE by 0.8-1.0 mm (all results reported as 95% confidence intervals), while skew or TPS deformation improved the mean TRE by <0.1 mm. The mean misalignment was 1.1-1.9^° (angle) and 0.9-1.3 mm (depth). Using isotropic scaling, the front face constraint raised the mean TRE by 0.6-0.8 mm.

Conclusions:

For sub-millimeter accuracy, 3D reconstruction models should not constrain histology images to the tissue slice front faces and should be flexible enough to model isotropic scaling.

Prostate histopathology: learning tissue component histograms for cancer detection and classification

Radical prostatectomy is performed on approximately 40% of men with organ-confined prostate cancer. Pathologic information obtained from the prostatectomy specimen provides important prognostic information and guides recommendations for adjuvant treatment. The current pathology protocol in most centers involves primarily qualitative assessment. In this paper, we describe and evaluate our system for automatic prostate cancer detection and grading on hematoxylin & eosin-stained tissue images. Our approach is intended to address the dual challenges of large data size and the need for high-level tissue information about the locations and grades of tumors. Our system uses two stages of AdaBoost-based classification. The first provides high-level tissue component labeling of a superpixel image partitioning. The second uses the tissue component labeling to provide a classification of cancer versus noncancer, and low-grade versus high-grade cancer. We evaluated our system using 991 sub-images extracted from digital pathology images of 50 whole-mount tissue sections from 15 prostatectomy patients. We measured accuracies of 90% and 85% for the cancer versus noncancer and high-grade versus low-grade classification tasks, respectively. This system represents a first step toward automated cancer quantification on prostate digital histopathology imaging, which could pave the way for more accurately informed postprostatectomy patient care.

Distinguishing radiation fibrosis from tumour recurrence after stereotactic ablative radiotherapy (SABR) for lung cancer: a quantitative analysis of CT density changes

BACKGROUND

For patients treated with stereotactic ablative radiotherapy (SABR) for early-stage non-small cell lung cancer, benign computed tomography (CT) changes due to radiation-induced lung injury (RILI) can be difficult to differentiate from recurrence. We measured the utility of CT image feature analysis in differentiating RILI from recurrence, compared to Response Evaluation Criteria In Solid Tumours (RECIST).

MATERIALS AND METHODS

Twenty-two patients with 24 lesions treated with SABR were selected (11 with recurrence, 13 with substantial RILI). On each follow-up CT, consolidative changes and ground glass opacities (GGO) were contoured. For each lesion, contoured regions were analysed for mean and variation in Hounsfield units (HU), 3D volume, and RECIST size during follow-up.

RESULTS

One hundred and thirty-six CT scans were reviewed, with a median imaging follow-up of 26 months. The 3D volume and RECIST measures of consolidative changes could significantly distinguish recurrence from RILI, but not until 15 months post-SABR; mean volume at 15 months [all values ± 95% confidence interval (CI)] of 30.1 ± 19.3 cm(3) vs. 5.1 ± 3.6 cm(3) (p = 0.030) and mean RECIST size at 15 months of 4.34 ± 1.13 cm vs. 2.63 ± 0.84 cm (p = 0.028) respectively for recurrence vs. RILI. At nine months post-SABR, patients with recurrence had significantly higher-density consolidative changes (mean at nine months of -96.4 ± 32.7 HU vs. -143.2 ± 28.4 HU for RILI; p = 0.046). They also had increased variability of HU, an image texture metric, measured as the standard deviation (SD) of HU, in the GGO areas (SD at nine months of 210.6 ± 14.5 HU vs. 175.1 ± 18.7 HU for RILI; p = 0.0078).

CONCLUSIONS

Quantitative changes in mean HU and GGO textural analysis have the potential to distinguish RILI from recurrence as early as nine months post-SABR, compared to 15 months with RECIST and 3D volume. If validated, this approach could allow for earlier detection and salvage of recurrence, and result in fewer unnecessary investigations of benign RILI.

The impact of registration accuracy on imaging validation study design: A novel statistical power calculation

Novel imaging modalities are pushing the boundaries of what is possible in medical imaging, but their signal properties are not always well understood. The evaluation of these novel imaging modalities is critical to achieving their research and clinical potential. Image registration of novel modalities to accepted reference standard modalities is an important part of characterizing the modalities and elucidating the effect of underlying focal disease on the imaging signal. The strengths of the conclusions drawn from these analyses are limited by statistical power. Based on the observation that in this context, statistical power depends in part on uncertainty arising from registration error, we derive a power calculation formula relating registration error, number of subjects, and the minimum detectable difference between normal and pathologic regions on imaging, for an imaging validation study design that accommodates signal correlations within image regions. Monte Carlo simulations were used to evaluate the derived models and test the strength of their assumptions, showing that the model yielded predictions of the power, the number of subjects, and the minimum detectable difference of simulated experiments accurate to within a maximum error of 1% when the assumptions of the derivation were met, and characterizing sensitivities of the model to violations of the assumptions. The use of these formulae is illustrated through a calculation of the number of subjects required for a case study, modeled closely after a prostate cancer imaging validation study currently taking place at our institution. The power calculation formulae address three central questions in the design of imaging validation studies: (1) What is the maximum acceptable registration error? (2) How many subjects are needed? (3) What is the minimum detectable difference between normal and pathologic image regions?