Melanoma is the most lethal form of skin cancer in the world. To improve the accuracy of diagnosis, quantitative imaging approaches have been investigated. While most quantitative methods focus on the surface of skin lesions via hand-crafted imaging features, in this work, we take a machine-learning approach where abstract quantitative imaging features are learned to model physiological traits. In doing so, we investigate skin cancer detection via computational modeling of two major physiological features of melanoma namely eumelanin and hemoglobin concentrations from dermal images. This was done via employing a non-linear random forest regression model to leverage the plethora of quantitative features from dermal images to build the model. The proposed method was validated by separability test applied to clinical images. The results showed that the proposed method outperforms state-of-the-art techniques on predicting the concentrations of the skin cancer physiological features in dermal images (i.e., eumelanin and hemoglobin).

The simultaneous capture of imaging data at multiple wavelengths across the electromagnetic spectrum is highly challenging, requiring complex and costly multispectral image devices. In this study, we investigate the feasibility of simultaneous multispectral imaging using conventional image sensors with color filter arrays via a novel comprehensive framework for numerical demultiplexing of the color image sensor measurements. A numerical forward model characterizing the formation of sensor measurements from light spectra hitting the sensor is constructed based on a comprehensive spectral characterization of the sensor. A numerical demultiplexer is then learned via non-linear random forest modeling based on the forward model. Given the learned numerical demultiplexer, one can then demultiplex simultaneously-acquired measurements made by the color image sensor into reflectance intensities at discrete selectable wavelengths, resulting in a higher resolution reflectance spectrum. Experimental results demonstrate the feasibility of such a method for the purpose of simultaneous multispectral imaging.

Radiomics has shown considerable promise as a new, emerging approach to computer-aided cancer screening. However, the idea of adopting radiomics for melanoma screening has not been previously explored, with clinical screening relying solely on visual assessment of skin lesion, and thus suffers from low sensitivity and specificity. In this study, a dermal radiomics framework is proposed for computer-aided screening of melanoma, with the aim of improving screening accuracy. A radiomic sequencer is designed to generate radiomic sequences consisting of 367 dermal radiomic features based on extracted physiological biomarkers from dermatological imaging data. The extracted dermal radiomic sequences were then employed to classify benign and malignant melanoma via non-linear random forest classification, and showed superior results in terms of sensitivity, specificity and accuracy when compared to the-state-of-the-art feature models for melanoma classification.

The current computer-aided melanoma diagnostic technique employs the features that extracted from the surface level of skin, and any under-skin information is ignored. Since the colour formation of lesion is vastly influenced by the concentration of melanin, the quantitative features of spatial heterogeneity of melanin concentration was proposed. To quantify the spatial heterogeneity of eumelanin and pheomelanin concentrations, the concentration maps were divided into two and the Earth mover's distance between the concentration clusters on both sides was measured. The proposed features were evaluated on 206 dermatological images (119 melanoma cases, 87 benign cases) using a cross-validation scheme. The results show that adopting spatial heterogeneity of melanin concentrations improved the sensitivity, specificity, and accuracy of diagnosing melanoma.

Traditional methods for early detection of melanoma rely upon a dermatologist to visually assess a skin lesion using the ABCDE (Asymmetry, Border irregularity, Color variegation, Diameter, Evolution) criteria before confirmation can be done through biopsy by a pathologist. However, this visual assessment strategy taken by dermatologists is hampered by clinician subjectivity and suffers from low sensitivity. Computer-aided diagnostic methods based on dermatological photographs are being developed to aid in the melanoma diagnosis process, but most of these methods rely only on superficial, topographic features that can be limiting in characterizing melanoma. In this work, a hybrid feature model is introduced for characterizing skin lesions that combines low-level and high-level features, and augments them with a set of physiological features extracted from dermatological photographs using a nearest-neighbor nonlinear model to improve classification performance. The physiological features extracted from the lesion for the proposed hybrid feature model include those based on: i) eumelanin concentrations, ii) pheomelanin concentrations, and iii) blood oxygen saturation. The proposed hybrid feature model was evaluated on 206 dermatological photographs of skin lesions (119 confirmed melanoma cases, 87 confirmed non-melanoma cases) using a cross validation scheme. The experimental results show that the proposed hybrid feature model, with integrated physiological features, provided improved sensitivity, specificity, precision and accuracy for the purpose of melanoma classification.

Intervertebral disc herniation is a very common disorder and contrast-enhanced computed tomography (CECT) is one of the imaging modalities for studying the causes of intervertebral disc herniation and its potential link as a mechanical source of pain. However, streak artefacts caused by the contrast agent reduce the quality of the reconstructed image. We therefore propose a novel image reconstruction technique for reducing streak artefacts in CECT images of the intervertebral disc. The technique identifies the contrast agent-affected region in projection space using a multi-scale segmentation algorithm, which is followed by reconstruction via Markov-chain Monte Carlo estimation. The results were compared with two existing artefact-reducing techniques (non-iterative and iterative), and the proposed method showed an improvement on signal-to-noise ratio (53.1 dB) while non-iterative and iterative approaches yielded 26.5 and 48.4 dB, respectively. The proposed image reconstruction technique can reduce streak artefacts on CECT images of intervertebral disc herniation and it can be extended to other streak artefacts caused by the contrast agent on computed tomography images.

Introduction Diffusion weighted imaging (DWI) has gained significant attention for prostate cancer imaging as it was found to be highly useful for tumor localization. The most common uses of DWI for prostate cancer analysis are for constructing apparent diffusion coefficient (ADC) maps, as well as for obtaining computed high b-value images with b-values greater than 1500 s/mm. As both the computation of ADC maps and computed high b-value images are derived from a set of DWI acquisitions across multiple b-values, any misalignments between the DWI acquisitions can result in degradation and inaccuracies in the derived images. To better mitigate the issue of misalignment, image registration of DWI acquisitions across multiple b-values is a necessary step. However, to the best of our knowledge, DWI co-registration across multiple b-values has not been well explored. Therefore, we propose a novel method to co-register high and low b-value DWI acquisitions together for improved ADC map computation and better visualization and localization of the prostate tumour. Method For registering DWI acquisitions, some important challenges include: i) deformation due to imaging parameters, ii) low SNR in high b-value DWI acquisitions, and iii) movement of patients. To overcome these issues, the proposed method adapts b-spline based non-linear registration by introducing a new hybrid similarity metric. The hybrid metric employs the optimized combination of statistical likelihood (SL) and cross correlation (CC) to adaptively accommodate for differences between DWI acquisitions at different b-values, where acquisitions with more similar b-values can be better handled using SL while the acquisitions with large b-value differences can be better handled by CC. Given that the target DWI acquisition (fb) with b-value (b) and the reference DWI acquisition (fref), the optimal transformation (T) is computed as: argmax ∈ ∈ ∈ ∈ where α , , and is the probability of given .

The current diagnostic technique for melanoma solely relies on the surface level of skin and under-skin information is neglected. Since physiological features of skin such as melanin are closely related to development of melanoma, the non-linear physiological feature extraction model based on random forest regression is proposed. The proposed model characterizes the concentration of eumelanin and pheomelanin from standard camera images or dermoscopic images, which are conventionally used for diagnosis of melanoma. For the validation, the phantom study and the separability test using clinical images were conducted and compared against the state-of-the art non-linear and linear feature extraction models. The results showed that the proposed model outperformed other comparing models in phantom and clinical experiments. Promising results show that the quantitative characterization of skin features, which is provided by the proposed method, can allow dermatologists and clinicians to make a more accurate and improved diagnosis of melanoma.

Improving the spatial resolution of Optical Coherence Tomography (OCT) images is important for the visualization and analysis of small morphological features in biological tissue such as blood vessels, membranes, cellular layers, etc. In this paper, we propose a novel reconstruction approach to obtaining super-resolved OCT tomograms from multiple lower resolution images. The proposed Multi-Penalty Conditional Random Field (MPCRF) method combines four different penalty factors (spatial proximity, first and second order intensity variations, as well as a spline-based smoothness of fit) into the prior model within a Maximum A Posteriori (MAP) estimation framework. Test carried out in retinal OCT images illustrate the effectiveness of the proposed MPCRF reconstruction approach in terms of spatial resolution enhancement, as compared to previously published super resolved image reconstruction methods. Visual assessment of the MPCRF results demonstrate the potential of this method in better preservation of fine details and structures of the imaged sample, as well as retaining the sharpness of biological tissue boundaries while reducing the effects of speckle noise inherent to OCT. Quantitative evaluation using imaging metrics such as Signal-to-Noise Ratio (SNR), Contrast to Noise Ratio (CNR), Equivalent Number of Looks (ENL), and Edge Preservation Parameter show significant visual quality improvement with the MPCRF approach. Therefore, the proposed MPCRF reconstruction approach is an effective tool for enhancing the spatial resolution of OCT images without the necessity for significant imaging hardware modifications.

Low-dose computed tomography (CT) reduces radiation exposure but decreases signal-to-noise ratio (SNR) and diagnostic capabilities. Noise compensation can improve SNR so low-dose CT can provide valuable information for diagnosis without risking patient radiation exposure. In this study, a novel noise-compensated CT reconstruction method that uses spatially adaptive Monte–Carlo sampling to produce noise-compensated reconstructions is investigated. By adapting to local noise statistics, a non-parametric estimation of the noise-free image is computed that successfully handles non-stationary noise found in low-dose CT images. Using phantom and real low-dose CT images, effective noise suppression is shown to be accomplished while maintaining structures and details.

Estimating heart movement and morphological changes during robot-assisted coronary artery bypass graft interventions C.A. Linte, D.S. Cho, M. Carias, D. Pace, J. Moore, C. Wedlake, D. Bainbridge, B. Kiaii,T. M. Peters Imaging Research Laboratories, Robarts Research Institute, London, Canada Biomedical Engineering, University of Western Ontario, London, Canada Schulich School of Medicine and Dentistry, University of Western Ontario, London, Canada Canadian Surgical Technologies and Advanced Robotics, London, Canada

PURPOSE:Although robot-assisted coronary artery bypass grafting (RA-CABG) has gained more acceptance worldwide, its success still depends on the surgeon's experience and expertise, and the conversion rate to full sternotomy is in the order of 15%-25%. One of the reasons for conversion is poor pre-operative planning, which is based solely on pre-operative computed tomography (CT) images. In this paper, the authors propose a technique to estimate the global peri-operative displacement of the heart and to predict the intra-operative target vessel location, validated via both an in vitro and a clinical study.METHODS:As the peri-operative heart migration during RA-CABG has never been reported in the literatures, a simple in vitro validation study was conducted using a heart phantom. To mimic the clinical workflow, a pre-operative CT as well as peri-operative ultrasound images at three different stages in the procedure (Stage(0)-following intubation; Stage(1)-following lung deflation; and Stage(2)-following thoracic insufflation) were acquired during the experiment. Following image acquisition, a rigid-body registration using iterative closest point algorithm with the robust estimator was employed to map the pre-operative stage to each of the peri-operative ones, to estimate the heart migration and predict the peri-operative target vessel location. Moreover, a clinical validation of this technique was conducted using offline patient data, where a Monte Carlo simulation was used to overcome the limitations arising due to the invisibility of the target vessel in the peri-operative ultrasound images.RESULTS:For the in vitro study, the computed target registration error (TRE) at Stage(0), Stage(1), and Stage(2) was 2.1, 3.3, and 2.6 mm, respectively. According to the offline clinical validation study, the maximum TRE at the left anterior descending (LAD) coronary artery was 4.1 mm at Stage(0), 5.1 mm at Stage(1), and 3.4 mm at Stage(2).CONCLUSIONS:The authors proposed a method to measure and validate peri-operative shifts of the heart during RA-CABG. In vitro and clinical validation studies were conducted and yielded a TRE in the order of 5 mm for all cases. As the desired clinical accuracy imposed by this procedure is on the order of one intercostal space (10-15 mm), our technique suits the clinical requirements. The authors therefore believe this technique has the potential to improve the pre-operative planning by updating peri-operative migration patterns of the heart and, consequently, will lead to reduced conversion to conventional open thoracic procedures.

MOTIVATION:Medical imaging and its application in interventional guidance has revolutionized the development of minimally invasive surgical procedures leading to reduced patient trauma, fewer risks, and shorter recovery times. However, a frequently posed question with regard to an image guidance system is "how accurate is it?" On one hand, the accuracy challenge can be posed in terms of the tolerable clinical error associated with the procedure; on the other hand, accuracy is bound by the limitations of the system's components, including modeling, patient registration, and surgical instrument tracking, all of which ultimately impact the overall targeting capabilities of the system. METHODS:While these processes are not unique to any interventional specialty, this paper discusses them in the context of two different cardiac image guidance platforms: a model-enhanced ultrasound platform for intracardiac interventions and a prototype system for advanced visualization in image-guided cardiac ablation therapy. RESULTS:Pre-operative modeling techniques involving manual, semi-automatic and registration-based segmentation are discussed. The performance and limitations of clinically feasible approaches for patient registration evaluated both in the laboratory and in the operating room are presented. Our experience with two different magnetic tracking systems for instrument and ultrasound transducer localization is reported. Ultimately, the overall accuracy of the systems is discussed based on both in vitro and preliminary in vivo experience. CONCLUSION:While clinical accuracy is specific to a particular patient and procedure and vastly dependent on the surgeon's experience, the system's engineering limitations are critical to determine whether the clinical requirements can be met.

OBJECTIVE:: For robot-assisted coronary artery bypass graft interventions, surgeons typically use a preoperative thoracic computed tomography scan of the patient to plan the procedure. However, the cardiac anatomy is prone to changes induced perioperatively in the effort to access the heart and surgical targets, which, in turn, may invalidate the initial plan. This article presents a method to estimate the perioperative heart migration, information which can be further exploited to refine the preoperative surgical plan.METHODS:: Tracked transesophageal ultrasound images of four patients' hearts were acquired at each stage in the procedure: before lung deflation, after lung deflation, and after both lung deflation and CO2 thoracic insufflation. Anatomic features of interest-the mitral and aortic valves-were identified from each dataset, and their movement between the different procedure stages was recorded and used to estimate the global heart displacement. Moreover, the local morphology of the features of interest was investigated to provide insight on the extent of the deformation the heart has undergone during the workflow.RESULTS:: The study suggested that the heart does undergo substantial displacement-on the order of 10 to 15 mm in each direction (axial, coronal, and sagittal) after lung deflation and CO2 thoracic insufflation. However, no significant differences (P > 0.1) were observed in the morphologic characteristics of the features of interest across the multiple workflow stages, suggesting that local deformations occur at a much smaller scale compared with the global migration.CONCLUSIONS:: The quantification of the perioperatively induced changes is critical to track the displacement of the heart and surgical targets. The recorded migration patterns should not be ignored but rather be used to update the surgical plan to better suit the intraoperative environment.

Prior to performing a robot-assisted coronary artery bypass grafting procedure, a pre-operative computed tomography scan is used to assess patient candidacy and to identify the location of the target vessel. The surgeon then determines the optimal port locations to ensure proper reach to the target with the robotic instruments, while assuming that the heart does not undergo any significant changes between the pre- and intra-operative stages. However, the peri-operative workflow itself leads to changes in heart position and consequently the intra-operative target vessel location. As such, the pre-operative plan must be adequately updated to adjust the target vessel location to better suit the intraoperative condition. Here we propose a technique to predict the position of the peri-operative target vessel location with approximately 3.5 mm RMS accuracy. We believe this technique will potentially reduce the rate of conversion of robot-assisted procedures to traditional open-chest surgery due to poor planning.

Image-guided interventions rely on the common assumption that pre-operative information can depict intraoperative morphology with sufficient accuracy. Nevertheless, in the context of minimally invasive cardiac therapy delivery, this assumption loses ground; the heart is a soft-tissue organ prone to changes induced during access to the heart and especially intracardiac targets. In addition to its clinical value for cardiac interventional guidance and assistance with the image-and model-to-patient registration, here we show how ultrasound imaging may be used to estimate changes in the heart position and morphology of structures of interest at different stages in the procedure. Using a magnetically tracked 2D transesophageal echocardiography transducer, we acquired in vivo images of the heart at different stages during the procedural workflow of common minimally invasive cardiac procedures, including robot-assisted coronary artery bypass grafting, mitral valve replacement/repair, or model-enhanced US-guided intracardiac interventions, all in the coordinate system of the tracking system. Anatomical features of interest (mitral and aortic valves) used to register the pre-operative anatomical models to the intraoperative coordinate frame were identified from each dataset. This information allowed us to identify the global position of the heart and also characterize the valvular structures at various peri-operative stages, in terms of their orientation, size, and geometry. Based on these results, we can estimate the differences between the pre- and intra-operative anatomical features, their effect on the model-to-subject registration, and also identify the need to update or optimize any pre-operative surgical plan to better suit the intra-operative procedure workflow.

Prostate cancer is a frequently occurring disease and is the second leading cause of cancer-related deaths of men in the United States. Current treatments have proved inadequate in curing or controlling prostate cancer, and a search for agents for the management of this disease is urgently needed. Survivin plays an important role in both progression of castration-resistant prostate cancer and resistance to chemotherapy. Altered expression of survivin in prostate cancer cells is associated with cancer progression, drug/radiation resistance, poor prognosis, and short patient survival. In the present study, the authors performed a cell-based rapid screen of the Prestwick Chemical Library consisting of 1120 Food and Drug Administration-approved compounds with known safety and bioavailability in humans to identify potential inhibitors of survivin and anticancer agents for prostate cancer. Sanguinarine, a benzophenanthridine alkaloid derived primarily from the bloodroot plant, was identified as a novel inhibitor of survivin that selectively kills prostate cancer cells over "normal" prostate epithelial cells. The authors found that sanguinarine inhibits survivin protein expression through protein degradation via the ubiquitin-proteasome system. Sanguinarine induces apoptosis and inhibits growth of human prostate cancer cells and in vivo tumor formation. Administration of sanguinarine, beginning 3 days after ectopic implantation of DU145 human prostate cancer cells, reduces both tumor weight and volume. In addition, sanguinarine sensitized paclitaxel-mediated growth inhibition and apoptosis, offering a potential therapeutic strategy for overcoming taxol resistance. These results suggest that sanguinarine may be developed as an agent either alone or in combination with taxol for treatment of prostate cancer overexpressing survivin.