Looking for the Right Diagnosis? Cardiovascular Magnetic Resonance Imaging Can Help Differentiate Cardiomyopathies.

Differentiating cardiomyopathies is a common clinical quandary in cardiology. Getting the right diagnosis is important for guiding patient management and providing prognosis. Incorrect or uncertain diagnoses can lead to further unnecessary investigations and/or treatment decisions applied inappropriately, which can have consequences for both the patient and health care costs. Cardiovascular magnetic resonance (CMR) imaging offers strength here due to its precision and breadth in assessing cardiac function and tissue characterisation. This review aims to raise awareness among cardiologists and physicians of the important insights provided by CMR-insights that can improve diagnosis and guide management, as well as aid in risk stratification, in different cardiomyopathies.

Cardiovascular Magnetic Resonance Parametric Mapping Techniques: Clinical Applications and Limitations.

PURPOSE OF REVIEW: Parametric mapping represents a significant innovation in cardiovascular magnetic resonance (CMR) tissue characterisation, allowing the quantification of myocardial changes based on changes on T1, T2 and T2* relaxation times and extracellular volume (ECV). Its clinical use is rapidly expanding, but it requires availability of dedicated equipment as well as expertise in image acquisition and analysis. This review focuses on the principles of CMR parametric mapping, its current clinical applications, important limitations, as well as future directions of this technique in cardiovascular medicine. RECENT FINDINGS: There is increasing evidence that CMR parametric mapping techniques provide accurate diagnostic and prognostic tools that can be applied to and support the clinical management of patients with a range of cardiovascular disease. The unique capability of CMR myocardial tissue characterisation in cardiovascular diseases has further expanded by the introduction of parametric mapping. Its use in clinical practice presents opportunities but has also limitations.

Cardiovascular Magnetic Resonance of Myocardial Fibrosis, Edema, and Infiltrates in Heart Failure.

Cardiac magnetic resonance (CMR) imaging is a unique imaging modality, which provides accurate noninvasive tissue characterization. Various CMR sequences can be utilized to identify and quantify patterns of myocardial edema, fibrosis, and infiltrates, which are important determinants for diagnosis and prognostication of heart failure. This article describes available methods of tissue characterization imaging applied in CMR. The presence and patterns of abnormal tissue characterization are related to common etiologies of heart failure and the techniques employed to demonstrate this. CMR provides the opportunity to identify the etiology of heart failure based on the recognition of different patterns of myocardial abnormalities.

Real-time cardiovascular magnetic resonance T1 and extracellular volume fraction mapping for tissue characterisation in aortic stenosis.

BACKGROUND: Myocardial fibrosis is a major determinant of outcome in aortic stenosis (AS). Novel fast real-time (RT) cardiovascular magnetic resonance (CMR) mapping techniques allow comprehensive quantification of fibrosis but have not yet been compared against standard techniques and histology. METHODS: Patients with severe AS underwent CMR before (n = 110) and left ventricular (LV) endomyocardial biopsy (n = 46) at transcatheter aortic valve replacement (TAVR). Midventricular short axis (SAX) native, post-contrast T1 and extracellular volume fraction (ECV) maps were generated using commercially available modified Look-Locker Inversion recovery (MOLLI) (native: 5(3)3, post-contrast: 4(1)3(1)2) and RT single-shot inversion recovery Fast Low-Angle Shot (FLASH) with radial undersampling. Focal late gadolinium enhancement was excluded from T1 and ECV regions of interest. ECV and LV mass were used to calculate LV matrix volumes. Variability and agreements were assessed between RT, MOLLI and histology using intraclass correlation coefficients, coefficients of variation and Bland Altman analyses. RESULTS: RT and MOLLI derived ECV were similar for midventricular SAX slice coverage (26.2 vs. 26.5, p = 0.073) and septal region of interest (26.2 vs. 26.5, p = 0.216). MOLLI native T1 time was in median 20 ms longer compared to RT (p < 0.001). Agreement between RT and MOLLI was best for ECV (ICC > 0.91), excellent for post-contrast T1 times (ICC > 0.81) and good for native T1 times (ICC > 0.62). Diffuse collagen volume fraction by biopsies was in median 7.8%. ECV (RT r = 0.345, p = 0.039; MOLLI r = 0.40, p = 0.010) and LV matrix volumes (RT r = 0.45, p = 0.005; MOLLI r = 0.43, p = 0.007) were the only parameters associated with histology. CONCLUSIONS: RT mapping offers fast and sufficient ECV and LV matrix volume calculation in AS patients. ECV and LV matrix volume represent robust and universally comparable parameters with associations to histologically assessed fibrosis and may emerge as potential targets for clinical decision making.

Cardiovascular Magnetic Resonance Imaging of Inherited Heart Conditions.

Imaging modalities are central to diagnosis and prognostication of confirmed or suspected inherited cardiomyopathies. The availability and use of cardiovascular magnetic resonance imaging (CMR) to supplement traditional modalities has increased substantially and has several advantages over traditional imaging techniques. CMR is unique in its ability to easily acquire images in any plane. Moreover, advances in CMR sequences have begun to enable characterisation of the myocardium without the need for invasive biopsy and has provided a major step forward in the understanding of inherited heart disease pathology and genotype-phenotype interactions. This review summarises the current role of CMR in inherited cardiomyopathies depending on their genotype and phenotype status, using arrhythmogenic right ventricular dysplasia/cardiomyopathy and hypertrophic cardiomyopathy as prototypical examples.

Diagnostic value of MRI-based 3D texture analysis for tissue characterisation and discrimination of low-grade chondrosarcoma from enchondroma: a pilot study.

OBJECTIVES: To explore the diagnostic value of MRI-based 3D texture analysis to identify texture features that can be used for discrimination of low-grade chondrosarcoma from enchondroma. METHODS: Eleven patients with low-grade chondrosarcoma and 11 patients with enchondroma were retrospectively evaluated. Texture analysis was performed using mint Lesion: Kurtosis, entropy, skewness, mean of positive pixels (MPP) and uniformity of positive pixel distribution (UPP) were obtained in four MRI sequences and correlated with histopathology. The Mann-Whitney U-test and receiver operating characteristic (ROC) analysis were performed to identify most discriminative texture features. Sensitivity, specificity, accuracy and optimal cut-off values were calculated. RESULTS: Significant differences were found in four of 20 texture parameters with regard to the different MRI sequences (p<0.01). The area under the ROC curve values to discriminate chondrosarcoma from enchondroma were 0.876 and 0.826 for kurtosis and skewness in contrast-enhanced T1 (ceT1w), respectively; in non-contrast T1, values were 0.851 and 0.822 for entropy and UPP, respectively. The highest discriminatory power had kurtosis in ceT1w with a cut-off ≥3.15 to identify low-grade chondrosarcoma (82 % sensitivity, 91 % specificity, accuracy 86 %). CONCLUSION: MRI-based 3D texture analysis might be able to discriminate low-grade chondrosarcoma from enchondroma by a variety of texture parameters. KEY POINTS: • MRI texture analysis may assist in differentiating low-grade chondrosarcoma from enchondroma. • Kurtosis in the contrast-enhanced T1w has the highest power of discrimination. • Tools provide insight into tumour characterisation as a non-invasive imaging biomarker.

Tissue shrinkage in microwave thermal ablation: comparison of three commercial devices.

PURPOSE: To evaluate, characterise and compare the extent of tissue shrinkage induced from three different commercial microwave ablation devices, and to elucidate the mechanism behind the distinctive performances obtained. MATERIALS AND METHODS: Microwave ablation (N = 152) was conducted with three different commercial devices on cubes of ex vivo liver (10-40 ± 2 mm/side) embedded in agar phantoms. 50-60 W was applied for 1-10 min duration. Pre- and post-ablation dimensions of the samples, as well as the extent of carbonisation and coagulation were measured and correlated. ANOVA was performed to evaluate statistical significance. RESULTS: For all devices, logarithmic correlations with time were observed for both tissue shrinkage (R2 = 0.84-1.00) and induced carbonisation (R2 = 0.73-0.99) radially to the antenna axis. Along the longitudinal axis of the antenna, for two of the devices shrinkage did not appreciably change with time (p > 0.05), yet carbonisation increased linearly (R2 = 0.57-0.94). For the third fully internally-cooled device, both carbonisation and shrinkage showed logarithmic trends (R2 = 0.85-0.98 and R2 = 0.78-0.94, respectively) based upon delayed carbonisation appearing only 5 min into ablation and onward. For all devices, non-uniform shrinkage was noted within the coagulated area increasing from the boundary of the ablated area (14%) to the limit of carbonisation (39%) in a linear fashion (R2 = 0.88) Conclusions: Microwave ablation device construction can alter the extent of post-ablation coagulation and tissue shrinkage. Given that tissue shrinkage in the coagulated area shows non-uniform behaviour, observed differences can be attributed in part to the applicator cooling system that alters the ablation temperature profile.

Experimental testing combined with inverse-FE for mechanical characterisation of penile tissues.

Erectile dysfunction (ED) predominantly affects men in their 40-70s and can lead to poor quality of life. One option for ED treatment is surgical implantation of an inflatable penile prosthesis (IPP). However, they can be associated with negative outcomes including infection, migration or fibrosis. To improve outcomes, the interaction between the IPP device and surrounding tissues needs further investigation and this could be achieved using pre-clinical testbeds, but they need to be informed by extensive tissue testing. In this study, an experimental approach is adopted to characterise the mechanics of horse penile tissue and establish a testing protocol for penile tissue. The whole penis segments were tested in plate compression tests to obtain whole penis behaviour which is necessary for validation of a pre-clinical testbed, whilst tensile and compression tests were performed on individual penile tissues, namely corpus cavernosa and tunica albuginea. The second part of the paper deals with the development of a computational model employing an inverse finite element approach to estimate the material parameters of each tissue layer. These material parameters are in good agreement with the experimental results obtained from the individual tissue layers and whole organ tissue tests. This paper presents the first study proposing realistic nonlinear elastic material parameters for penile tissues and offers a validated testbed for IPPs. STATEMENT OF SIGNIFICANCE: Erectile Dysfunction (ED) affects over half the male population aged 40-70 potentially leading to poor quality of life. Patients not responding to conventional treatments of ED, are advised to use penile prostheses which can create an erection using implanted inflatable cylinders. A significant drawback of such prostheses, however, is the substantial tissue damage they can induce during their usage. Preclinical testbeds, including computational and bench-top models, could offer an efficient means of improving device designs to mitigate this damage but such testbeds require extensive knowledge of penile tissue properties. In this study, the authors determine penile tissue mechanics and apply an inverse FE approach to characterise the penile material properties required to validate preclinical models of the penis.

Real-time haptic characterisation of Hunt-Crossley model based on radial basis function neural network for contact environment.

Dynamic soft tissue characterisation is an important element in robotic minimally invasive surgery. This paper presents a novel method by combining neural network with recursive least square (RLS) estimation for dynamic soft tissue characterisation based on the nonlinear Hunt-Crossley (HC) model. It develops a radial basis function neural network (RBFNN) to compensate for the error caused by natural logarithmic factorisation (NLF) of the HC model for dynamic RLS estimation of soft tissue properties. The RBFNN weights are estimated according to the maximum likelihood principle to evaluate the probability distribution of the neural network modelling residual. Further, by using the linearisation error modelled by RBFNN to compensate for the linearised HC model, an RBFNN-based RLS algorithm is developed for dynamic soft tissue characterisation. Simulation and experimental results demonstrate that the proposed method can effectively model the natural logarithmic linearisation error, leading to improved accuracy for RLS estimation of the HC model parameters.

Is location a significant parameter in the layer dependent dissection properties of the aorta?

Proper characterisation of biological tissue is key to understanding the effect of the biomechanical environment in the physiology and pathology of the cardiovascular system. Aortic dissection in particular is a prevalent and sometimes fatal disease that still lacks a complete comprehension of its progression. Its development and outcome, however, depend on the location in the vessel. Dissection properties of arteries are frequently studied via delamination tests, such as the T-peel test and the mixed-mode peel test. So far, a study that performs both tests throughout different locations of the aorta, as well as dissecting several interfaces, is missing. This makes it difficult to extract conclusions in terms of vessel heterogeneity, as a standardised experimental procedure cannot be assured for different studies in literature. Therefore, both dissection tests have been here performed on healthy porcine aortas, dissecting three interfaces of the vessels, i.e., the intima-media, the media-adventitia and the media within itself, considering different locations of the aorta, the ascending thoracic aorta (ATA), the descending thoracic aorta and the infrarenal abdominal aorta (IAA). Significant differences were found for both, layers and location. In particular, dissection forces in the ATA were the highest and the separation of the intima-media interface required significantly the lowest force. Moreover, dissection in the longitudinal direction of the vessel generally required more force than in the circumferential one. These results emphasise the need to characterise aortic tissue considering the specific location and dissected layer of the vessel.

Machine learning-enabled quantitative ultrasound techniques for tissue differentiation.

PURPOSE: Quantitative ultrasound (QUS) infers properties about tissue microstructure from backscattered radio-frequency ultrasound data. This paper describes how to implement the most practical QUS parameters using an ultrasound research system for tissue differentiation. METHODS: This study first validated chicken liver and gizzard muscle as suitable acoustic phantoms for human brain and brain tumour tissues via measurement of the speed of sound and acoustic attenuation. A total of thirteen QUS parameters were estimated from twelve samples, each using data obtained with a transducer with a frequency of 5-11 MHz. Spectral parameters, i.e., effective scatterer diameter and acoustic concentration, were calculated from the backscattered power spectrum of the tissue, and echo envelope statistics were estimated by modelling the scattering inside the tissue as a homodyned K-distribution, yielding the scatterer clustering parameter α and the structure parameter κ. Standard deviation and higher-order moments were calculated from the echogenicity value assigned in conventional B-mode images. RESULTS: The k-nearest neighbours algorithm was used to combine those parameters, which achieved 94.5% accuracy and 0.933 F1-score. CONCLUSION: We were able to generate classification parametric images in near-real-time speed as a potential diagnostic tool in the operating room for the possible use for human brain tissue characterisation.

Cardiovascular magnetic resonance characterisation of anthracycline cardiotoxicity in adults with normal left ventricular ejection fraction.

BACKGROUND: Anthracycline therapy may lead to changes in cardiac structure and function not detectable by solely evaluating left ventricular ejection fraction (LVEF). OBJECTIVES: We hypothesized that cardiovascular magnetic resonance (CMR) would identify structural and functional myocardial abnormalities in anthracycline-treated cancer survivors with normal LVEF, compared to a matched control population. METHODS: Forty-five cancer survivors (56 ± 16 yrs., 60% female) with normal LVEF (59.5 ± 4.1%) were studied a median of 11 months (range 3-36) following administration of 237 ± 83 mg/m2 anthracycline, and compared with forty-five healthy control subjects of similar age and sex (53 ± 16 yrs., 60% female) with normal LVEF (60.8 ± 2.4%) using 1.5 T CMR. RESULTS: Significantly smaller indexed left ventricular mass (45.6 ± 8.7 vs 50.3 ± 10.1 g/m2, p = 0.02) and indexed myocardial cell volume (30.5 ± 5.7 vs 34.8 ± 7.2 ml/m2, p = 0.002) were evident in cancer survivors and the latter was inversely associated with cumulative anthracycline dose (r = -0.31, p = 0.02). Surrogate CMR markers of myocardial fibrosis were significantly increased in cancer survivors (native myocardial T1: 1021 ± 40 vs 996 ± 35 ms, p = 0.002; extracellular volume: 29.5 ± 4.5 vs 27.4 ± 2.3%, p = 0.006). CMR-derived feature-tracking global longitudinal strain (GLS) was significantly impaired in cancer survivors (2D GLS -18.3 ± 2.6 vs -20.0 ± 2.0%, p < 0.001; 3D GLS -14.5 ± 2.3 vs -16.4 ± 2.6%, p < 0.001). Parameters exhibited good to excellent (ICC = 0.86-0.98) inter- and intra-observer reproducibility. CONCLUSIONS: Anthracycline-treated cancer survivors with normal LVEF have significant perturbations of LV mass, myocardial cell volume, native myocardial T1, ECV, CMR-derived 2D and 3D GLS, compared to controls, with good to excellent levels of inter- and intra-observer reproducibility.

Mechanical, compositional and morphological characterisation of the human male urethra for the development of a biomimetic tissue engineered urethral scaffold.

This study addresses a crucial gap in the literature by characterising the relationship between urethral tissue mechanics, composition and gross structure. We then utilise these data to develop a biomimetic urethral scaffold with physical properties that more accurately mimic the native tissue than existing gold standard scaffolds; small intestinal submucosa (SIS) and urinary bladder matrix (UBM). Nine human urethra samples were mechanically characterised using pressure-diameter and uniaxial extension testing. The composition and gross structure of the tissue was determined using immunohistological staining. A pressure stiffening response is observed during the application of intraluminal pressure. The elastic and viscous tissue responses to extension are free of regional or directional variance. The elastin and collagen content of the tissue correlates significantly with tissue mechanics. Building on these data, a biomimetic urethral scaffold was fabricated from collagen and elastin in a ratio that mimics the composition of the native tissue. The resultant scaffold is comprised of a dense inner layer and a porous outer layer that structurally mimic the submucosa and corpus spongiosum layers of the native tissue, respectively. The porous outer layer facilitated more uniform cell infiltration relative to SIS and UBM when implanted subcutaneously (p < 0.05). The mechanical properties of the biomimetic scaffold better mimic the native tissue compared to SIS and UBM. The tissue characterisation data presented herein paves the way for the development of biomimetic urethral grafts, and the novel scaffold we develop demonstrates positive findings that warrant further in vivo evaluation.

Clinical Translation of Three-Dimensional Scar, Diffusion Tensor Imaging, Four-Dimensional Flow, and Quantitative Perfusion in Cardiac MRI: A Comprehensive Review.

Cardiovascular magnetic resonance (CMR) imaging is a versatile tool that has established itself as the reference method for functional assessment and tissue characterisation. CMR helps to diagnose, monitor disease course and sub-phenotype disease states. Several emerging CMR methods have the potential to offer a personalised medicine approach to treatment. CMR tissue characterisation is used to assess myocardial oedema, inflammation or thrombus in various disease conditions. CMR derived scar maps have the potential to inform ablation therapy-both in atrial and ventricular arrhythmias. Quantitative CMR is pushing boundaries with motion corrections in tissue characterisation and first-pass perfusion. Advanced tissue characterisation by imaging the myocardial fibre orientation using diffusion tensor imaging (DTI), has also demonstrated novel insights in patients with cardiomyopathies. Enhanced flow assessment using four-dimensional flow (4D flow) CMR, where time is the fourth dimension, allows quantification of transvalvular flow to a high degree of accuracy for all four-valves within the same cardiac cycle. This review discusses these emerging methods and others in detail and gives the reader a foresight of how CMR will evolve into a powerful clinical tool in offering a precision medicine approach to treatment, diagnosis, and detection of disease.

Are all left bundle branch blocks the same? Myocardial mechanical implications by cardiovascular magnetic resonance.

AIMS: Left bundle branch block (LBBB) is usually associated with structural myocardial diseases progressively leading to left ventricular (LV) dysfunction. We sought to determine the mechanical implications of LBBB (as defined based on Strauss' criteria) by Cardiovascular Magnetic Resonance (CMR). METHOD AND RESULTS: We included consecutive patients referred to CMR to assess the structural cause of LBBB. CMR scans consisted of cine, stress perfusion, and late gadolinium enhancement (LGE) sequences. Myocardial deformation was assessed by tissue tracking analysis; LGE was quantified using the full width at half maximum method. We included 86 patients [63% male, 70 years (60-72)] with mean QRS duration 150 ± 13 msec. A structural disease was identified on CMR in 53% of patients (ischemic heart disease, IHD, 31%; non-ischemic heart disease, NIHD, 22%), while LBBB-related septal dyssynchrony (SD) was the only abnormality in 47%. LGE was found in 42% of patients. LVEF and myocardial deformation were impaired. Despite similar ECG characteristics, myocardial strain differed significantly between IHD, NIHD and SD patients, and patients with SD showed less impaired myocardial deformation. Indexed LV end-systolic volume and LGE extent were independently associated with impaired strain. CONCLUSIONS: Patients with LBBB show different structural and mechanical properties, and LGE extent has an unfavourable effect on myocardial mechanics.

Correlation between calcium, water contents and ultrasonographic appearance of atherosclerotic lesions of carotid artery lesions.

BACKGROUND: We tested the hypothesis whether there is a correlation between the echogenicity and calcium and water contents of carotid plaques. PATIENTS AND METHODS: Ninety carotid befurcations from 45 deceased patients were removed during autopsy. Thirty-four plaques were categorized as homogenous echolucent (HEL), homogenous echogenic (HEG) and heterogenous (HE) plaques based on premortem B-mode image. Water content was expressed in % of wet weight. Ca was determined by proton-induced X-ray emission and expressed in ppm. Relative optical density of the B-mode images was analyzed offline using a computer program. RESULTS: HEL plaques had lower Ca content (medians and IQRs: 6,145 [4,465-6,536 ppm]) compared to HEG (74,100 [15,300-1,44,500-ppm]), P ≤ 0.001). HE plaques showed an intermediate calcium content (7,310 [4,840-9,920 ppm]) that was statistically not different from echolucent plaques. Water content of HEG plaques was statistically not different from HEL and HE (HEG:53.5 [35.5-64%], HEL: 73.5 [69.7-78.5%], HE: 70.6 [67.4-73.9%]). HEG plaques had the highest relative optical densities (196 [188-217%]). HEL and HE had similar relative optical densities (HEL: 176 [164-187%], HE: 164 [144-188%], respectively). A significant positive correlation was found between the Ca content and relative optical density of plaques. CONCLUSIONS: Echogenicity of carotid plaques increases along with their calcium content. Water content may be an important factor in differentiation of different plaques.

Ultrasound tissue characterisation of the superficial digital flexor tendons in juvenile Thoroughbred racehorses during early race training.

BACKGROUND: Injuries to the superficial digital flexor tendon (SDFT) are one of the leading causes of Thoroughbred (TB) wastage. Increasingly, the aim is to prevent injury rather than treat it. Conventional ultrasonography is not sufficiently sensitive to accurately monitor tendon and predict injury. Ultrasound tissue characterisation (UTC) is a relatively new technique, which improves tendon characterisation by providing a 3-dimensional (3D) SDFT reconstruction and objective calculation of fibre alignment by classifying fibres into one of 4 echo-types. OBJECTIVES: To report a reference range of echo-types in a population of normal juvenile TB racehorses. It was hypothesised that: UTC would be easy to use on juvenile TB racehorses in a field setting; that results would be repeatable; that the UTC would demonstrate a physiologic response of the tendon and, finally, that the technique would allow monitoring of the SDFT for early detection of degenerative changes. STUDY DESIGN: Prospective longitudinal cohort pilot study. METHODS: Thirty-two TB yearling racehorses were recruited. UTC measurements of bilateral forelimb SDFTs were taken every 60-90 days. The proportion of 4 echo-types were quantified as a relative percentage at specific zones over the length of the SDFT. Relationships were assessed by paired T tests or Wilcoxon signed rank tests. RESULTS: Mean percentage for echo-type I fibres were >85%; echo-type II fibres were <15%, with negligible echo-type III and IV. Significant right to left limb, zonal, and temporal differences in echo-type were identified. MAIN LIMITATIONS: No control group of untrained horses, limiting ability to differentiate whether findings were training-related as opposed to age-related changes. CONCLUSIONS: Changes in SDFT characterisation over the first 6 months of training were identified. UTC may provide useful objective information when assessing juvenile SDFTs. The Summary is available in Spanish - see Supporting Information.

Context aware decision support in neurosurgical oncology based on an efficient classification of endomicroscopic data.

PURPOSE: Probe-based confocal laser endomicroscopy (pCLE) enables in vivo, in situ tissue characterisation without changes in the surgical setting and simplifies the oncological surgical workflow. The potential of this technique in identifying residual cancer tissue and improving resection rates of brain tumours has been recently verified in pilot studies. The interpretation of endomicroscopic information is challenging, particularly for surgeons who do not themselves routinely review histopathology. Also, the diagnosis can be examiner-dependent, leading to considerable inter-observer variability. Therefore, automatic tissue characterisation with pCLE would support the surgeon in establishing diagnosis as well as guide robot-assisted intervention procedures. METHODS: The aim of this work is to propose a deep learning-based framework for brain tissue characterisation for context aware diagnosis support in neurosurgical oncology. An efficient representation of the context information of pCLE data is presented by exploring state-of-the-art CNN models with different tuning configurations. A novel video classification framework based on the combination of convolutional layers with long-range temporal recursion has been proposed to estimate the probability of each tumour class. The video classification accuracy is compared for different network architectures and data representation and video segmentation methods. RESULTS: We demonstrate the application of the proposed deep learning framework to classify Glioblastoma and Meningioma brain tumours based on endomicroscopic data. Results show significant improvement of our proposed image classification framework over state-of-the-art feature-based methods. The use of video data further improves the classification performance, achieving accuracy equal to 99.49%. CONCLUSIONS: This work demonstrates that deep learning can provide an efficient representation of pCLE data and accurately classify Glioblastoma and Meningioma tumours. The performance evaluation analysis shows the potential clinical value of the technique.

The effect of load on Achilles tendon structure in novice runners.

OBJECTIVES: To observe the changes in Achilles tendon structure in novice runners, with loading prescriptions of 100% body weight compared to 20% body weight. DESIGN: Randomised crossover. METHODS: Twenty novice runners participated in two separate running bouts spaced 14days apart, one of high load at 100% body weight, and one of low load at 20% body weight. Tendon structure was measured by ultrasonographic tissue characterisation on 6 occasions; immediately prior to each run, 2 and 7days after each run. RESULTS: The interaction effect of time and condition was not found to be significant for echotypes I-IV [Wald chi-square=2.8, d.f.=2, P=0.247; Wald chi-square=2.888, d.f.=2, P=0.236; Wald chi-square=1.385, d.f.=2, P=0.5; Wald chi-square=4.19, d.f.=2, P=0.123], respectively. A significant effect of time was found for echotypes III [Wald chi-square=6.785, d.f.=2, P=0.0.034] and IV [Wald chi-square=7.491, d.f.=2, P=0.0.024]. CONCLUSIONS: The decrease in echotypes III and IV suggest that moderate loads can be applied to the Achilles tendon without compromising tendon structure. Low to moderate loads may be beneficial in the management of Achilles tendinopathy. Further studies should focus on protocols with higher loading and/or repetitive loading in athletic populations with and without Achilles tendinopathy to assess any differences in tendon structure.

Ultrasound elastography and ultrasound tissue characterisation for tendon evaluation.

Ultrasound elastography (UE) and ultrasound tissue characterisation (UTC) are two newer modes of ultrasound (US) which have begun to attract scientific interests as ways to improve tendon characterisation. These modes of US show early promise in improved diagnostic accuracy, prediction of at-risk tendons and prognostication capability beyond conventional grey-scale US. Here, we provide a review of the literature on UE and UTC for Achilles, patellar and rotator cuff tendons. The translational potential of this article: The present literature indicates that UE and UTC could potentially increase the clinician's ability to accurately diagnose the extent of tendon pathology, including preclinical injury.