This PDF file contains the editorial “Special Section Guest Editorial:Radiomics and Imaging Genomics: Quantitative Imaging for Precision Medicine” for JMI Vol. 2 Issue 04

Image heterogeneity metrics such as textural features are an active area of research for evaluating clinical outcomes with positron emission tomography (PET) imaging and other modalities. However, the effects of stochastic image acquisition noise on these metrics are poorly understood. We performed a simulation study by generating 50 statistically independent PET images of the NEMA IQ phantom with realistic noise and resolution properties. Heterogeneity metrics based on gray-level intensity histograms, co-occurrence matrices, neighborhood difference matrices, and zone size matrices were evaluated within regions of interest surrounding the lesions. The impact of stochastic variability was evaluated with percent difference from the mean of the 50 realizations, coefficient of variation and estimated sample size for clinical trials. Additionally, sensitivity studies were performed to simulate the effects of patient size and image reconstruction method on the quantitative performance of these metrics. Complex trends in variability were revealed as a function of textural feature, lesion size, patient size, and reconstruction parameters. In conclusion, the sensitivity of PET textural features to normal stochastic image variation and imaging parameters can be large and is feature-dependent. Standards are needed to ensure that prospective studies that incorporate textural features are properly designed to measure true effects that may impact clinical outcomes.

Intensive international efforts are underway toward phenotyping the entire mouse genome by modifying all its ≈25,000 genes one-by-one for comparative studies. A workload of this scale has triggered numerous studies harnessing image informatics for the identification of morphological defects. However, existing work in this line primarily rests on abnormality detection via structural volumetrics between wild-type and gene-modified mice, which generally fails when the pathology involves no severe volume changes, such as ventricular septal defects (VSDs) in the heart. Furthermore, in embryo cardiac phenotyping, the lack of relevant work in embryonic heart segmentation, the limited availability of public atlases, and the general requirement of manual labor for the actual phenotype classification after abnormality detection, along with other limitations, have collectively restricted existing practices from meeting the high-throughput demands. This study proposes, to the best of our knowledge, the first fully automatic VSD classification framework in mouse embryo imaging. Our approach leverages a combination of atlas-based segmentation and snake evolution techniques to derive the segmentation of heart ventricles, where VSD classification is achieved by checking whether the left and right ventricles border or overlap with each other. A pilot study has validated our approach at a proof-of-concept level and achieved a classification accuracy of 100% through a series of empirical experiments on a database of 15 images.

Current computer-aided diagnosis (CAD) models for determining pulmonary nodule malignancy characterize nodule shape, density, and border in computed tomography (CT) data. Analyzing the lung parenchyma surrounding the nodule has been minimally explored. We hypothesize that improved nodule classification is achievable by including features quantified from the surrounding lung tissue. To explore this hypothesis, we have developed expanded quantitative CT feature extraction techniques, including volumetric Laws texture energy measures for the parenchyma and nodule, border descriptors using ray-casting and rubber-band straightening, histogram features characterizing densities, and global lung measurements. Using stepwise forward selection and leave-one-case-out cross-validation, a neural network was used for classification. When applied to 50 nodules (22 malignant and 28 benign) from high-resolution CT scans, 52 features (8 nodule, 39 parenchymal, and 5 global) were statistically significant. Nodule-only features yielded an area under the ROC curve of 0.918 (including nodule size) and 0.872 (excluding nodule size). Performance was improved through inclusion of parenchymal (0.938) and global features (0.932). These results show a trend toward increased performance when the parenchyma is included, coupled with the large number of significant parenchymal features that support our hypothesis: the pulmonary parenchyma is influenced differentially by malignant versus benign nodules, assisting CAD-based nodule characterizations.

Combining imaging and genetic information to predict disease presence and progression is being codified into an emerging discipline called “radiogenomics.” Optimal evaluation methodologies for radiogenomics have not been well established. We aim to develop a decision framework based on utility analysis to assess predictive models for breast cancer diagnosis. We garnered Gail risk factors, single nucleotide polymorphisms (SNPs), and mammographic features from a retrospective case-control study. We constructed three logistic regression models built on different sets of predictive features: (1) Gail, (2) Gail + Mammo, and (3) Gail + Mammo + SNP. Then we generated receiver operating characteristic (ROC) curves for three models. After we assigned utility values for each category of outcomes (true negatives, false positives, false negatives, and true positives), we pursued optimal operating points on ROC curves to achieve maximum expected utility of breast cancer diagnosis. We performed McNemar’s test based on threshold levels at optimal operating points, and found that SNPs and mammographic features played a significant role in breast cancer risk estimation. Our study comprising utility analysis and McNemar’s test provides a decision framework to evaluate predictive models in breast cancer risk estimation.

We analyzed the spatial diversity of tumor habitats, regions with distinctly different intensity characteristics of a tumor, using various measurements of habitat diversity within tumor regions. These features were then used for investigating the association with a 12-month survival status in glioblastoma (GBM) patients and for the identification of epidermal growth factor receptor (EGFR)-driven tumors. T1 postcontrast and T2 fluid attenuated inversion recovery images from 65 GBM patients were analyzed in this study. A total of 36 spatial diversity features were obtained based on pixel abundances within regions of interest. Performance in both the classification tasks was assessed using receiver operating characteristic (ROC) analysis. For association with 12-month overall survival, area under the ROC curve was 0.74 with confidence intervals [0.630 to 0.858]. The sensitivity and specificity at the optimal operating point (threshold=0.5) on the ROC were 0.59 and 0.75, respectively. For the identification of EGFR-driven tumors, the area under the ROC curve (AUC) was 0.85 with confidence intervals [0.750 to 0.945]. The sensitivity and specificity at the optimal operating point (threshold=0.166) on the ROC were 0.76 and 0.83, respectively. Our findings suggest that these spatial habitat diversity features are associated with these clinical characteristics and could be a useful prognostic tool for magnetic resonance imaging studies of patients with GBM.

Genomic and radiomic imaging profiles of invasive breast carcinomas from The Cancer Genome Atlas and The Cancer Imaging Archive were integrated and a comprehensive analysis was conducted to predict clinical outcomes using the radiogenomic features. Variable selection via LASSO and logistic regression were used to select the most-predictive radiogenomic features for the clinical phenotypes, including pathological stage, lymph node metastasis, and status of estrogen receptor (ER), progesterone receptor (PR), and human epidermal growth factor receptor 2 (HER2). Cross-validation with receiver operating characteristic (ROC) analysis was performed and the area under the ROC curve (AUC) was employed as the prediction metric. Higher AUCs were obtained in the prediction of pathological stage, ER, and PR status than for lymph node metastasis and HER2 status. Overall, the prediction performances by genomics alone, radiomics alone, and combined radiogenomics features showed statistically significant correlations with clinical outcomes; however, improvement on the prediction performance by combining genomics and radiomics data was not found to be statistically significant, most likely due to the small sample size of 91 cancer cases with 38 radiomic features and 144 genomic features.

Laser interstitial thermal therapy (LITT) has recently emerged as a new treatment modality for cancer pain management that targets the cingulum (pain center in the brain) and has shown promise over radio frequency (RF)-based ablation, due to magnetic resonance image (MRI) guidance that allows for precise ablation. Since laser ablation for pain management is currently exploratory and is only performed at a few centers worldwide, its short- and long-term effects on the cingulum are currently unknown. Traditionally, treatment effects for neurological conditions are evaluated by monitoring changes in intensities and/or volume of the ablation zone on post-treatment Gadolinium-contrast T1-w (Gd-T1) MRI. However, LITT introduces subtle localized changes corresponding to tissues response to treatment, which may not be appreciable on visual inspection of volumetric or intensity changes. Additionally, different MRI protocols [Gd-T1, T2w, gradient echo sequence (GRE), fluidattenuated inversion recovery (FLAIR)] are known to capture complementary diagnostic information regarding the patient’s response to treatment; the utility of these MRI protocols has so far not been investigated to evaluate early and localized response to LITT treatment in the context of neuropathic cancer pain. In this work, we present the first attempt at (a) examining early treatment-related changes on a per-voxel basis via quantitative comparison of computer-extracted texture descriptors across pre- and post-LITT multiparametric (MP-MRI) (Gd-T1, T2w, GRE, FLAIR), subtle microarchitectural texture changes that may not be appreciable on original MR intensities or volumetric differences, and (b) investigating the efficacy of different MRI protocols in accurately capturing immediate post-treatment changes reflected (1) within and (2) outside the ablation zone. A retrospective cohort of four patient studies comprising pre- and immediate (24 h) post-LITT 3 Tesla Gd-T1, T2w, GRE, and FLAIR acquisitions was considered. Our quantitative approach first involved intensity standardization to allow for grayscale MR intensities acquired pre- and post-LITT to have a fixed tissue-specific meaning within the same imaging protocol, the same body region, and within the same patient. An affine registration was then performed on individual post-LITT MRI protocols to a reference MRI protocol pre-LITT. A total of 78 computerized texture features (co-occurrence matrix homogeneity, neighboring gray-level dependence matrix, Gabor) are then extracted from pre- and post-LITT MP-MRI on a per-voxel basis. Quantitative, voxelwise comparison of the changes in MRI texture features between pre- and post-LITT MRI indicate that (a) Gabor texture features at specific orientations were highly sensitive as well as specific in predicting subtle microarchitectural changes within and around the ablation zone pre- and post-LITT, (b) FLAIR was identified as the most sensitive MRI protocol in identifying early treatment changes yielding a normalized percentage change of 360% within the ablation zone relative to its pre-LITT value, and (c) GRE was identified as the most sensitive MRI protocol in quantifying changes outside the ablation zone post-LITT. Our preliminary results thus indicate potential for noninvasive computerized MP-MRI features over volumetric features in determining localized microarchitectural early focal treatment changes post-LITT for neuropathic cancer pain treatment.

This paper presents an imaging-genomic pipeline to derive three-dimensional intra-tumor heterogeneity features from contrast-enhanced CT images and correlates them with gene mutation status. The pipeline has been demonstrated using CT scans of patients with clear cell renal cell carcinoma (ccRCC) from The Cancer Genome Atlas. About 15% of ccRCC cases reported have BRCA1-associated protein 1 (BAP1) gene alterations that are associated with high tumor grade and poor prognosis. We hypothesized that BAP1 mutation status can be detected using computationally derived image features. The molecular data pertaining to gene mutation status were obtained from the cBioPortal. Correlation of the image features with gene mutation status was assessed using the Mann-Whitney-Wilcoxon rank-sum test. We also used the random forests classifier in the Waikato Environment for Knowledge Analysis software to assess the predictive ability of the computationally derived image features to discriminate cases with BAP1 mutations for ccRCC. Receiver operating characteristics were obtained using a leave-one-out-cross-validation procedure. Our results show that our model can predict BAP1 mutation status with a high degree of sensitivity and specificity. This framework demonstrates a methodology for noninvasive disease biomarker detection from contrast-enhanced CT images.

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.

The purpose of this study is to investigate the utility of obtaining “core samples” of regions in CT volume scans for extraction of radiomic features. We asked four readers to outline tumors in three representative slices from each phase of multiphasic liver CT images taken from 29 patients (1128 segmentations) with hepatocellular carcinoma. Core samples were obtained by automatically tracing the maximal circle inscribed in the outlines. Image features describing the intensity, texture, shape, and margin were used to describe the segmented lesion. We calculated the intraclass correlation between the features extracted from the readers’ segmentations and their core samples to characterize robustness to segmentation between readers, and between human-based segmentation and core sampling. We conclude that despite the high interreader variability in manually delineating the tumor (average overlap of 43% across all readers), certain features such as intensity and texture features are robust to segmentation. More importantly, this same subset of features can be obtained from the core samples, providing as much information as detailed segmentation while being simpler and faster to obtain.

Thanks to the current advances in nanoscience, molecular biochemistry, and x-ray detector technology, x-ray fluorescence computed tomography (XFCT) has been considered for molecular imaging of probes containing high atomic number elements, such as gold nanoparticles. The commonly used XFCT imaging performed with K-shell x rays appears to have insufficient imaging sensitivity to detect the low gold concentrations observed in small animal studies. Low energy fluorescence L-shell x rays have exhibited higher signal-to-background ratio and appeared as a promising XFCT mode with greatly enhanced sensitivity. The aim of this work was to experimentally demonstrate the feasibility of L-shell XFCT imaging and to assess its achievable sensitivity. We built an experimental L-shell XFCT imaging system consisting of a miniature x-ray tube and two spectrometers, a silicon drift detector (SDD), and a CdTe detector placed at ±120  deg with respect to the excitation beam. We imaged a 28-mm-diameter water phantom with 4-mm-diameter Eppendorf tubes containing gold solutions with concentrations of 0.06 to 0.1% Au. While all Au vials were detectable in the SDD L-shell XFCT image, none of the vials were visible in the CdTe L-shell XFCT image. The detectability limit of the presented L-shell XFCT SDD imaging setup was 0.007% Au, a concentration observed in small animal studies.

Wide-angle x-ray scatter (WAXS) could potentially be used to diagnose ductal carcinoma in situ (DCIS) in breast biopsies. The regions of interest were assumed to consist of fibroglandular tissue and epithelial cells and the model assumed that biopsies with DCIS would have a higher concentration of the latter. The scattered number of photons from a 2-mm diameter column of tissue was simulated using a 110-kV beam and selectively added in terms of momentum transfer. For a 1-min exposure, specificities and sensitivities of unity were obtained for biopsies 2- to 20-mm thick. The impact of sample and tumor cell layer thicknesses was studied. For example, a biopsy erroneously estimated to be 8 mm would be correctly diagnosed if its actual thickness was between 7.3 and 8.7 mm. An 8-mm thick malignant biopsy can be correctly diagnosed provided the malignant cell layer thickness is <0.96  mm. WAXS methods could become a diagnostic tool for DCIS within breast biopsies.

Biplane fluoroscopy is used for dynamic in vivo three-dimensional motion analysis of various joints of the body. Cross-scatter between the two fluoroscopy systems may limit tracking accuracy. This study measured the magnitude and effects of cross-scatter in biplane fluoroscopic images. Four cylindrical phantoms of 4-, 6-, 8-, and 10-in. diameter were imaged at varying kVp levels to determine the cross-scatter fraction and contrast-to-noise ratio (CNR). Monte Carlo simulations quantified the effect of the gantry angle on the cross-scatter fraction. A cadaver foot with implanted beads was also imaged. The effect of cross-scatter on marker-based tracking accuracy was investigated. Results demonstrated that the cross-scatter fraction varied from 0.15 for the 4-in. cylinder to 0.89 for the 10-in. cylinder when averaged across kVp. The average change in CNR due to cross-scatter ranged from 5% to 36% CNR decreases for the 4- and 10-in. cylinders, respectively. In simulations, the cross-scatter fraction increased with the gantry angle for the 8- and 10-in. cylinders. Cross-scatter significantly increased static-tracking error by 15%, 25%, and 38% for the 6-, 8-, and 10-in. phantoms, respectively, with no significant effect for the foot specimen. The results demonstrated submillimeter marker-based tracking for a range of phantom sizes, despite cross-scatter degradation.

Head computed tomography (CT) plays an important role in the comprehensive evaluation of acute stroke. Photon-counting spectral detectors, as promising candidates for use in the next generation of x-ray CT systems, allow for assigning more weight to low-energy x-rays that generally contain more contrast information. Most importantly, the spectral information can be utilized to decompose the original set of energy-selective images into several basis function images that are inherently free of beam-hardening artifacts, a potential advantage for further improving the diagnosis accuracy. We are developing a photon-counting spectral detector for CT applications. The purpose of this work is to determine the optimal beam quality for material decomposition in two head imaging cases: nonenhanced imaging and K-edge imaging. A cylindrical brain tissue of 16-cm diameter, coated by a 6-mm-thick bone layer and 2-mm-thick skin layer, was used as a head phantom. The imaging target was a 5-mm-thick blood vessel centered in the head phantom. In K-edge imaging, two contrast agents, iodine and gadolinium, with the same concentration (5  mg/mL) were studied. Three parameters that affect beam quality were evaluated: kVp settings (50 to 130 kVp), filter materials (Z=13 to 83), and filter thicknesses [0 to 2 half-value layer (HVL)]. The image qualities resulting from the varying x-ray beams were compared in terms of two figures of merit (FOMs): squared signal-difference-to-noise ratio normalized by brain dose (SDNR²/BD) and that normalized by skin dose (SDNR²/SD). For nonenhanced imaging, the results show that the use of the 120-kVp spectrum filtered by 2 HVL copper (Z=29) provides the best performance in both FOMs. When iodine is used in K-edge imaging, the optimal filter is 2 HVL iodine (Z=53) and the optimal kVps are 60 kVp in terms of SDNR²/BD and 75 kVp in terms of SDNR²/SD. A tradeoff of 65 kVp was proposed to lower the potential risk of skin injuries if a relatively long exposure time is necessarily performed in the iodinated imaging. In the case of gadolinium imaging, both SD and BD can be minimized at 120 kVp filtered with 2 HVL thulium (Z=69). The results also indicate that with the same concentration and their respective optimal spectrum, the values of SDNR²/BD and SDNR²/SD in gadolinium imaging are, respectively, around 3 and 10 times larger than those in iodine imaging. However, since gadolinium is used in much lower concentrations than iodine in the clinic, iodine may be a preferable candidate for K-edge imaging.

While many approaches exist to segment retinal vessels in fundus photographs, only a limited number focus on the construction and disambiguation of arterial and venous trees. Previous approaches are local and/or greedy in nature, making them susceptible to errors or limiting their applicability to large vessels. We propose a more global framework to generate arteriovenous trees in retinal images, given a vessel segmentation. In particular, our approach consists of three stages. The first stage is to generate an overconnected vessel network, named the vessel potential connectivity map (VPCM), consisting of vessel segments and the potential connectivity between them. The second stage is to disambiguate the VPCM into multiple anatomical trees, using a graph-based metaheuristic algorithm. The third stage is to classify these trees into arterial or venous (A/V) trees. We evaluated our approach with a ground truth built based on a public database, showing a pixel-wise classification accuracy of 88.15% using a manual vessel segmentation as input, and 86.11% using an automatic vessel segmentation as input.

This paper aims to characterize the radiation dose and image quality (IQ) performance of a dental cone beam computed tomography (CBCT) unit over a range of fields of view (FOV). IQ and dose were measured using a Carestream 9300 dental CBCT. Phantoms were positioned in the FOV to imitate clinical positioning. IQ was assessed by scanning a SEDENTEXCT IQ phantom, and images were analyzed in ImageJ. Dose index 1 was obtained using a thimble ionization chamber and SEDENTEXCT DI phantom. Mean gray values agreed within 93.5% to 99.7% across the images, with pixel-to-pixel fluctuations of 6% to 12.5%, with poorer uniformity and increased noise for child protocols. CNR was fairly constant across FOVs, with higher CNR for larger patient settings. The measured limiting spatial resolution agreed well with 10% MTF and bar pattern measurements. Dose was reduced for smaller patient settings within a given FOV; however, smaller FOVs obtained with different acquisition settings did not necessarily result in reduced dose. The use of patient-specific acquisition settings decreased the radiation dose for smaller patients, with minimal impact on the IQ. The full set of IQ and dose measurements is reported to allow dental professionals to compare the different FOV settings for clinical use.

As the panoramic x-ray is the most common extraoral radiography in dentistry, segmentation of its anatomical structures facilitates diagnosis and registration of dental records. This study presents a fast and accurate method for automatic segmentation of mandible in panoramic x-rays. In the proposed four-step algorithm, a superior border is extracted through horizontal integral projections. A modified Canny edge detector accompanied by morphological operators extracts the inferior border of the mandible body. The exterior borders of ramuses are extracted through a contour tracing method based on the average model of mandible. The best-matched template is fetched from the atlas of mandibles to complete the contour of left and right processes. The algorithm was tested on a set of 95 panoramic x-rays. Evaluating the results against manual segmentations of three expert dentists showed that the method is robust. It achieved an average performance of <93% in Dice similarity, specificity, and sensitivity.

Laparoscopic surgery, which is one minimally invasive surgical technique that is now widely performed, is done by making a working space (pneumoperitoneum) by infusing carbon dioxide (CO₂) gas into the abdominal cavity. A virtual pneumoperitoneum method that simulates the abdominal wall and viscera motion by the pneumoperitoneum based on mass-spring-damper models (MSDMs) with mechanical properties is proposed. Our proposed method simulates the pneumoperitoneum based on MSDMs and Newton’s equations of motion. The parameters of MSDMs are determined by the anatomical knowledge of the mechanical properties of human tissues. Virtual CO₂ gas pressure is applied to the boundary surface of the abdominal cavity. The abdominal shapes after creation of the pneumoperitoneum are computed by solving the equations of motion. The mean position errors of our proposed method using 10 mmHg virtual gas pressure were 26.9±5.9  mm, and the position error of the previous method proposed by Kitasaka et al. was 35.6 mm. The differences in the errors were statistically significant (p<0.001, Student’s t-test). The position error of the proposed method was reduced from 26.9±5.9 to 23.4±4.5  mm using 30 mmHg virtual gas pressure. The proposed method simulated abdominal wall motion by infused gas pressure and generated deformed volumetric images from a preoperative volumetric image. Our method predicted abdominal wall deformation by just giving the CO₂ gas pressure and the tissue properties. Measurement of the visceral displacement will be required to validate the visceral motion.

An edge-enhancement technique using an interval type II fuzzy set that uses rank-ordered filter to enhance the edges of cellular images is proposed. When cellular images from any laboratory are digitized, scanned, and stored, some kind of degradation occurs, and directly using a rank-ordered filter may not produce clear edges. These images contain uncertainties, present in edges or boundaries of the image. Fuzzy sets that take into account these uncertainties may be a good tool to process these images. However, a fuzzy set sometimes does not produce better results. We used an interval type II fuzzy set, which considers the uncertainty in a different way. It considers the membership function in the fuzzy set as “fuzzy,” so the membership values lie within an interval range. A type II fuzzy set has upper and lower membership levels, and with the two levels, a new membership function is computed using Hamacher t-conorm. A new fuzzy image is formed. A rank-ordered filter is applied to the image to obtain an edge-enhanced image. The proposed method is compared with the existing methods visually and quantitatively using entropic method. Entropy of the proposed method is higher (0.4418) than the morphology method (0.2275), crisp method (0.3599), and Sobel method (0.2669), implying that the proposed method is better.

The thyroid is an endocrine gland that regulates metabolism. Thyroid image analysis plays an important role in both diagnostic radiology and radiation oncology treatment planning. Low tissue contrast of the thyroid relative to surrounding anatomic structures makes manual segmentation of this organ challenging. This work proposes a fully automated system for thyroid segmentation on CT imaging. Following initial thyroid segmentation with multiatlas joint label fusion, a random forest (RF) algorithm was applied. Multiatlas label fusion transfers labels from labeled atlases and warps them to target images using deformable registration. A consensus atlas solution was formed based on optimal weighting of atlases and similarity to a given target image. Following the initial segmentation, a trained RF classifier employed voxel scanning to assign class-conditional probabilities to the voxels in the target image. Thyroid voxels were categorized with positive labels and nonthyroid voxels were categorized with negative labels. Our method was evaluated on CT scans from 66 patients, 6 of which served as atlases for multiatlas label fusion. The system with independent multiatlas label fusion method and RF classifier achieved average dice similarity coefficients of 0.72±0.13 and 0.57±0.14, respectively. The system with sequential multiatlas label fusion followed by RF correction increased the dice similarity coefficient to 0.76±0.11 and improved the segmentation accuracy.

We propose an architecture with a remote phosphor-based modular and compact light-emitting diode (LED) light source in a noncontact dermoscope prototype for skin cancer screening. The spectrum and color temperature of the output light can easily and significantly be changed depending on spectral absorption characteristics of the tissues being imaged. The new system has several advantages compared to state-of-the-art phosphor converted ultrabright white LEDs, used in a wide range of medical imaging devices, which have a fixed spectrum and color temperature at a given operating point. In particular, the system can more easily be adapted to the requirements originating from different tissues in the human body, which have wavelength-dependent absorption and reflectivity. This leads to improved contrast for different kinds of imaged tissue components. The concept of such a lighting architecture can be vastly utilized in many other medical imaging devices including endoscopic systems.

Up to 25% of children who undergo brain tumor resection surgery in the posterior fossa develop posterior fossa syndrome (PFS). This syndrome is characterized by mutism and disturbance in speech. Our hypothesis is that there is a correlation between PFS and the occurrence of hypertrophic olivary degeneration (HOD) in structures within the posterior fossa, known as the inferior olivary nuclei (ION). HOD is exhibited as an increase in size and intensity of the ION on an MR image. Longitudinal MRI datasets of 28 patients were acquired consisting of pre-, intra-, and postoperative scans. A semiautomated segmentation process was used to segment the ION on each MR image. A full set of imaging features describing the first- and second-order statistics and size of the ION were extracted for each image. Feature selection techniques were used to identify the most relevant features among the MRI features, demographics, and data based on neuroradiological assessment. A support vector machine was used to analyze the discriminative features selected by a generative k-nearest neighbor algorithm. The results indicate the presence of hyperintensity in the left ION as the most diagnostically relevant feature, providing a statistically significant improvement in the classification of patients (p=0.01) when using this feature alone.

The early detection of melanoma is one of the greatest challenges in clinical practice of dermatology, and the reticular pattern is one of the most important dermoscopic structures to improve melanocytic lesion diagnosis. A texture-based approach is developed for the automatic detection of reticular patterns, whose output will assist clinical decision-making. Feature selection was based on the use of two algorithms by means of the classical graylevel co-occurrence matrix and Laws energy masks optimized on a set of 104 dermoscopy images. The AdaBoost (adaptive boosting) approach to machine learning was used within this strategy. Results suggest superiority of LEM for reticular pattern detection in dermoscopic images, achieving a sensitivity of 90.16% and a specificity of 86.67%. The use of automatic classification in dermoscopy to support clinicians is a strong tool to assist diagnosis; however, the use of automatic classification as a complementary tool in clinical routine requires algorithms with high levels of sensitivity and specificity. The results presented in this work will contribute to achieving this goal.

The goal of computer-assisted surgery is to provide the surgeon with guidance during an intervention, e.g., using augmented reality. To display preoperative data, soft tissue deformations that occur during surgery have to be taken into consideration. Laparoscopic sensors, such as stereo endoscopes, can be used to create a three-dimensional reconstruction of stereo frames for registration. Due to the small field of view and the homogeneous structure of tissue, reconstructing just one frame, in general, will not provide enough detail to register preoperative data, since every frame only contains a part of an organ surface. A correct assignment to the preoperative model is possible only if the patch geometry can be unambiguously matched to a part of the preoperative surface. We propose and evaluate a system that combines multiple smaller reconstructions from different viewpoints to segment and reconstruct a large model of an organ. Using graphics processing unit-based methods, we achieved four frames per second. We evaluated the system with in silico, phantom, ex vivo, and in vivo (porcine) data, using different methods for estimating the camera pose (optical tracking, iterative closest point, and a combination). The results indicate that the proposed method is promising for on-the-fly organ reconstruction and registration.

Laparoscopic surgery is a minimally invasive surgical technique where surgeons insert a small video camera into the patient’s body to visualize internal organs and use small tools to perform surgical procedures. However, the benefit of small incisions has a drawback of limited visualization of subsurface tissues, which can lead to navigational challenges in the delivering of therapy. Image-guided surgery uses the images to map subsurface structures and can reduce the limitations of laparoscopic surgery. One particular laparoscopic camera system of interest is the vision system of the daVinci-Si robotic surgical system (Intuitive Surgical, Sunnyvale, California). The video streams generate approximately 360 MB of data per second, demonstrating a trend toward increased data sizes in medicine, primarily due to higher-resolution video cameras and imaging equipment. Processing this data on a bedside PC has become challenging and a high-performance computing (HPC) environment may not always be available at the point of care. To process this data on remote HPC clusters at the typical 30 frames per second (fps) rate, it is required that each 11.9 MB video frame be processed by a server and returned within 1/30th of a second. The ability to acquire, process, and visualize data in real time is essential for the performance of complex tasks as well as minimizing risk to the patient. As a result, utilizing high-speed networks to access computing clusters will lead to real-time medical image processing and improve surgical experiences by providing real-time augmented laparoscopic data. We have performed image processing algorithms on a high-definition head phantom video (1920 × 1080 pixels) and transferred the video using a message passing interface. The total transfer time is around 53 ms or 19 fps. We will optimize and parallelize these algorithms to reduce the total time to 30 ms.

Beyond estimation of depot volumes, quantitative analysis of adipose tissue properties could improve understanding of how adipose tissue correlates with metabolic risk factors. We investigated whether the fat signal fraction (FSF) derived from quantitative fat–water magnetic resonance imaging (MRI) scans at 3.0 T correlates to CT Hounsfield units (HU) of the same tissue. These measures were acquired in the subcutaneous white adipose tissue (WAT) at the umbilical level of 21 healthy adult subjects. A moderate correlation exists between MRI- and CT-derived WAT values for all subjects, R²=0.54, p<0.0001, with a slope of −2.6, (95% CI [−3.3,−1.8]), indicating that a decrease of 1 HU equals a mean increase of 0.38% FSF. We demonstrate that FSF estimates obtained using quantitative fat–water MRI techniques correlate with CT HU values in subcutaneous WAT, and therefore, MRI-based FSF could be used as an alternative to CT HU for assessing metabolic risk factors.

IHE XDS-I profile proposes an architecture model for cross-enterprise medical image sharing, but there are only a few clinical implementations reported. Here, we investigate three pilot studies based on the IHE XDS-I profile to see whether we can use this architecture as a foundation for image sharing solutions in a variety of health-care settings. The first pilot study was image sharing for cross-enterprise health care with federated integration, which was implemented in Huadong Hospital and Shanghai Sixth People’s Hospital within the Shanghai Shen-Kang Hospital Management Center; the second pilot study was XDS-I–based patient-controlled image sharing solution, which was implemented by the Radiological Society of North America (RSNA) team in the USA; and the third pilot study was collaborative imaging diagnosis with electronic health-care record integration in regional health care, which was implemented in two districts in Shanghai. In order to support these pilot studies, we designed and developed new image access methods, components, and data models such as RAD-69/WADO hybrid image retrieval, RSNA clearinghouse, and extension of metadata definitions in both the submission set and the cross-enterprise document sharing (XDS) registry. We identified several key issues that impact the implementation of XDS-I in practical applications, and conclude that the IHE XDS-I profile is a theoretically good architecture and a useful foundation for medical image sharing solutions across multiple regional health-care providers.

This article presents an approach to biomedical image retrieval by mapping image regions to local concepts where images are represented in a weighted entropy-based concept feature space. The term “concept” refers to perceptually distinguishable visual patches that are identified locally in image regions and can be mapped to a glossary of imaging terms. Further, the visual significance (e.g., visualness) of concepts is measured as the Shannon entropy of pixel values in image patches and is used to refine the feature vector. Moreover, the system can assist the user in interactively selecting a region-of-interest (ROI) and searching for similar image ROIs. Further, a spatial verification step is used as a postprocessing step to improve retrieval results based on location information. The hypothesis that such approaches would improve biomedical image retrieval is validated through experiments on two different data sets, which are collected from open access biomedical literature.

Image-guided core needle biopsy is the current gold standard for breast cancer diagnosis. Microcalcifications, an important radiographic finding on mammography suggestive of early breast cancer such as ductal carcinoma in situ, are usually biopsied under stereotactic guidance. This procedure, however, is uncomfortable for patients and requires the use of ionizing radiation. It would be preferable to biopsy microcalcifications under ultrasound guidance since it is a faster procedure, more comfortable for the patient, and requires no radiation. However, microcalcifications cannot reliably be detected with the current standard ultrasound imaging systems. This study is motivated by the clinical need for real-time high-resolution ultrasound imaging of microcalcifications, so that biopsies can be accurately performed under ultrasound guidance. We have investigated how high-frequency ultrasound imaging can enable visualization of microstructures in ex vivo breast tissue biopsy samples. We generated B-mode images of breast tissue and applied the Nakagami filtering technique to help refine image output so that microcalcifications could be better assessed during ultrasound-guided core biopsies. We describe the preliminary clinical results of high-frequency ultrasound imaging of ex vivo breast biopsy tissue with microcalcifications and without Nakagami filtering and the correlation of these images with the pathology examination by hematoxylin and eosin stain and whole slide digital scanning.

The elastic properties of tissue are related to tissue composition and pathological changes. It has been observed that many pathological processes increase the elastic modulus of soft tissue compared to normal. Ultrasound compression elastography is a method of characterization of elastic properties that has been the focus of many research efforts in the last two decades. In medical radiology, compression elastography is provided as an additional tool with ultrasound B-mode in the existing scanners, and the combined features of elastography and echography act as a promising diagnostic method in breast cancer detection. However, the full capability of the ultrasound elastography technique together with B-mode has not been utilized by novice radiologists due to the nonavailability of suitable, appropriately designed tissue-mimicking phantoms. Since different commercially available ultrasound elastographic scanners follow their own unique protocols, training novice radiologists is becoming cumbersome. The main focus of this work is to develop a tissue-like agar-based phantom, which mimics breast tissue with common abnormal lesions like fibroadenoma and invasive ductal carcinoma in a clinically perceived way and compares the sonographic and elastographic appearances using different commercially available systems. In addition, the developed phantoms are simulated using the finite-element method, and ideal strain images are generated. Strain images from experiment and simulation are compared based on image contrast parameters, namely contrast transfer efficiency (CTE) and observed strain, and they are in good agreement. The strain image contrast of malignant inclusions is significantly improved compared to benign inclusions, and the trend of CTE is similar for all elastographic scanners under investigation.