Identifying patients with undiagnosed small intestinal neuroendocrine tumours in primary care using statistical and machine learning: model development and validation study.

BACKGROUND: Neuroendocrine tumours (NETs) are increasing in incidence, often diagnosed at advanced stages, and individuals may experience years of diagnostic delay, particularly when arising from the small intestine (SI). Clinical prediction models could present novel opportunities for case finding in primary care. METHODS: An open cohort of adults (18+ years) contributing data to the Optimum Patient Care Research Database between 1st Jan 2000 and 30th March 2023 was identified. This database collects de-identified data from general practices in the UK. Model development approaches comprised logistic regression, penalised regression, and XGBoost. Performance (discrimination and calibration) was assessed using internal-external cross-validation. Decision analysis curves compared clinical utility. RESULTS: Of 11.7 million individuals, 382 had recorded SI NET diagnoses (0.003%). The XGBoost model had the highest AUC (0.869, 95% confidence interval [CI]: 0.841-0.898) but was mildly miscalibrated (slope 1.165, 95% CI: 1.088-1.243; calibration-in-the-large 0.010, 95% CI: -0.164 to 0.185). Clinical utility was similar across all models. DISCUSSION: Multivariable prediction models may have clinical utility in identifying individuals with undiagnosed SI NETs using information in their primary care records. Further evaluation including external validation and health economics modelling may identify cost-effective strategies for case finding for this uncommon tumour.

COSMIC: the Catalogue Of Somatic Mutations In Cancer.

COSMIC, the Catalogue Of Somatic Mutations In Cancer (https://cancer.sanger.ac.uk) is the most detailed and comprehensive resource for exploring the effect of somatic mutations in human cancer. The latest release, COSMIC v86 (August 2018), includes almost 6 million coding mutations across 1.4 million tumour samples, curated from over 26 000 publications. In addition to coding mutations, COSMIC covers all the genetic mechanisms by which somatic mutations promote cancer, including non-coding mutations, gene fusions, copy-number variants and drug-resistance mutations. COSMIC is primarily hand-curated, ensuring quality, accuracy and descriptive data capture. Building on our manual curation processes, we are introducing new initiatives that allow us to prioritize key genes and diseases, and to react more quickly and comprehensively to new findings in the literature. Alongside improvements to the public website and data-download systems, new functionality in COSMIC-3D allows exploration of mutations within three-dimensional protein structures, their protein structural and functional impacts, and implications for druggability. In parallel with COSMIC's deep and broad variant coverage, the Cancer Gene Census (CGC) describes a curated catalogue of genes driving every form of human cancer. Currently describing 719 genes, the CGC has recently introduced functional descriptions of how each gene drives disease, summarized into the 10 cancer Hallmarks.

Septic Pylephlebitis in the Setting of COVID-19 Infection: A Case Report.

Portal vein thrombophlebitis is a rare complication that can occur in various hypercoagulable states, including COVID-19. We are presenting a 74-year-old female with a history of hypertension, diabetes, and lymphoma who contracted the COVID-19 infection and presented with persistent fever, leukocytosis, and mild epigastric tenderness. She developed hypotension, acute hypoxic respiratory failure, and worsening leukocytosis with bandemia and was diagnosed with portal vein thrombosis (PVT) and superior mesenteric vein thrombosis. The patient received broad-spectrum IV antibiotics and full anticoagulation therapy with heparin and was discharged on oral Warfarin after completing 14-day antibiotic therapy. She presented again with recurrent watery diarrhea, fever, abdominal pain, and fatigue and was diagnosed with pylephlebitis and multiple small liver abscesses. The patient was treated with antibiotics for six weeks and was discharged on warfarin, furosemide, and spironolactone with close outpatient follow-up. Prolonged fever in COVID-19 patients can indicate extensive thrombosis at unusual sites, which can lead to major morbidity and mortality in patients.

Community-Acquired Meningitis Complicated With Pyogenic Ventriculitis and Hydrocephalus in a Patient With Haematological Malignancy: A Case Report and Literature Review.

Meningitis, an infection of the meninges of the central nervous system (CNS), can advance quickly and carries a mortality rate reaching 30% among affected patients. It may become complicated by conditions such as hydrocephalus, ventriculitis, and cerebral abscess. Here, we describe a case of meningitis that was complicated by pyogenic ventriculitis and hydrocephalus in a patient with diffuse large B-cell lymphoma (DLBCL) who underwent chemotherapy and radiotherapy. The patient presented with acute change in mental status and high-grade fever, with few episodes of non-bloody vomiting. Blood culture and cerebrospinal fluid (CSF) culture grew Streptococcus pneumoniae, which was sensitive to ceftriaxone. CT scan of the head showed ventriculomegaly, pansinusitis, and a large left mastoid effusion. MRI of the brain showed layering in ventricles, hydrocephalus, and dural enhancement consistent with pachymeningitis. She was treated with ceftriaxone for 21 days with a meaningful outcome. She was discharged home with near-baseline mental capacity for further physical therapy.

The wall signal removal in Doppler ultrasound systems based on recursive PCA.

In Doppler ultrasound (US) systems, a high-pass filter is usually employed to remove the wall component from the blood flow signal. However, this will lead to the loss of information from the low velocity flow. In this paper, an algorithm based on the principal components analysis (PCA) is proposed, in which singular value decomposition (SVD) is used to extract the main component from the mixed signals. Furthermore, the recursive process is incorporated into the PCA method to improve the performance of wall signal removal. This approach and the traditional high-pass filtering one are, respectively, applied to analyze the computer-simulated in vitro and in vivo Doppler US signals. With the proposed method, the wall signal can be removed while a large portion of low-velocity blood signal remains. Comparison experiments show that this novel approach can satisfy the requirements of Doppler US system and is practicable under a broad range of measurement conditions. Because this algorithm is based on real data, it is currently applied to unidirectional signals.

Angle-independent estimation of maximum velocity through stenoses using vector Doppler ultrasound.

Categorisation for arterial stenoses treatment is determined primarily by the degree of occlusion, which is often estimated ultrasonically from blood velocity measurements. In current single-beam ultrasound (US) systems, this estimate can suffer from gross errors due to angle-dependence. The purpose of this study was to find out if an experimental dual-beam US system could reduce the angle-dependence of the velocity estimates. We compared four dual-beam velocity estimation algorithms on both a string phantom and straight tube wall-less flow phantoms incorporating symmetrical and asymmetrical stenoses from 0% to 91% by area. The estimated maximum velocity varied, on average, by 7.6% for beam-vessel angles from 40 degrees to 80 degrees. The fluctuation in the magnitude estimate was reduced by a factor of 2.6 using a hybrid single-dual-beam algorithm. We conclude that, when the true velocity lies in the scan plane, the dual-beam system reduces the angle-dependence and, thus, has the potential to improve categorisation of patients with arterial stenoses.

Lesion classification using skin patterning.

BACKGROUND/AIMS: The observation that skin line patterning tends to be disrupted by malignant but not non-malignant skin lesions suggests that this could be used as an aid to lesion differentiation. Since recognised differentiating features can be obtained from the simply-captured white light optical image, the possibility of using such images for skin pattern disruption detection is worth exploring. METHODS: The skin pattern has been extracted from optical images by high-pass filtering and profiles of local line strength variation with the angle estimated using a new consistent high-value profiling technique. The resultant profile images have been analysed using a novel region-based agglomerative clustering technique (mRAC) and also a local variance measurement. A measure based on the relationship between the classification results and an intensity-based segmentation was calculated, and this represented the disruption of the skin line patterning. RESULTS: A set of images containing a variety of histologically confirmed malignant and non-malignant lesions was analysed. The computed textural disruption figure was compared to both the histological diagnosis and to a visual estimate of patterning disruption for each image. It was demonstrated that lesion separation could be achieved by both analysis methods, with a good correlation with visual estimate of disruption and with mRAC providing the best performance. CONCLUSIONS: It was concluded that the acquisition and modelling of skin line patterning from clinical images of skin lesions had been successfully achieved and that the analysis of the resulting data provided an assessment of pattern disruption that is both consistent with visual inspection and effective in presenting information useful for discrimination between melanoma and benign naevi lesion examples.

Lumen pressure within obliquely insonated absorbent solid cylindrical shells with application to Doppler flow phantoms.

Flow phantoms used in medical ultrasound usually employ a plastic tube as a blood vessel mimic. These tubes often have acoustic properties differing significantly from the tissue and blood-mimicking media, which results in distortion of the acoustic pressure field within the tubes and, hence, of the Doppler flow spectra. Previous analyses of this problem have used some form of the infinite plate transmission coefficient, although at least one ray-based analysis has considered a cylindrical interface but with zero wall thickness. In this paper, we compare these approximate pressure fields with the exact solution for oblique incidence on a viscoelastic cylindrical shell at 5 MHz to find for which materials the plate approximation is valid. The shell has water both inside and outside, but it can be modified to use a different fluid inside and also to include absorption in either fluid. We find the plate approximation is reasonable for soft tubes such as the copolymer Cflex (Cole-Palmer, Niles, IL) but much less so for hard tubes such as polymethylmethacrylate (PMMA).

A simulation study of sample volume sensitivity for oblique pulsed finite beam insonation of Doppler ultrasound flow phantom cylindrical vessels.

Our previous analysis of the lumen pressure in Doppler ultrasound flow phantoms subject to continuous wave, infinite beam excitation is extended here to consider the pressure and Doppler sample volume complex sensitivity within a range of solid absorbent tubes typical of those used in Doppler ultrasound flow phantoms insonated with a focussed pulsed ultrasound beam. The beam may be incident on the cylindrical shell from any angle and with any offset from the shell axis. The examples considered are of a 5 MHz beam with a 6 dB lateral fullwidth of 1 mm at the focus and a transducer surface acceleration pulse with standard deviation of 1 micros propagating through 10 mm outer diameter, 8 mm inner diameter, Cflex, low-density polyethylene (LDPE), high-density polyethylene (HDPE), and polymethylmethacrylate (PMMA) shells surrounded by water at various beam-vessel angles. Our results confirm earlier analyses suggesting that PMMA, being less well matched to the surrounding media, causes much greater distortion of the sample volume sensitivity than Cflex.

Sample volume misregistration in linear array-based dual beam Doppler ultrasound systems.

Large velocity estimation errors can occur in dual beam Doppler ultrasound velocity measurement systems when there is left/right sample volume misregistration, particularly when the interbeam angle is small. Such misregistration will occur when there is tissue inhomogeneity. This is investigated for a typical type of inhomogeneity--a layer of fat--by calculating the amount of both angle and translation misregistration occurring in such a system realized using a single linear array transducer. The complex sample volume sensitivity is calculated using a modified time domain approach, combining the spatial impulse response method with ray tracing. The effects on these misregistrations of altering the aperture sizes and their relative positions on the array is then investigated to derive an improved aperture configuration for dual beam velocity estimation. Arrangements with transmit apertures wider than the receive apertures are shown to be preferable in this context.

Error propagation bounds in dual and triple beam vector Doppler ultrasound.

Measurement of the velocity components along two (three) directions enables the two (three)-dimensional velocity vector to be estimated exactly. However, in practical systems employing such multiple beam techniques, there will usually be errors in the measured velocity components along each beam, which will lead to errors in the estimated velocity magnitude and direction. This error propagation problem is analyzed in both two and three dimensions by decomposition of the velocity estimation matrix, and exact upper and lower bounds are derived for both the magnitude and angle bias as a function of the angle between the beams. The bias in triple beam systems is shown to have identical bounds to that in dual beam systems with an equivalent interbeam angle. It is found that small errors in the individual beam velocity components can be magnified in the final determination of velocity magnitude and angle. Plots are presented to assist system designers to specify the interbeam angle(s) to avoid gross velocity estimation errors.

Velocity bias and fluctuation in the standard dual beam Doppler reconstruction algorithm.

Bias and fluctuation of the standard velocity reconstruction algorithm for dual beam vector Doppler velocity estimation systems are analyzed; both magnitude and angle properties are considered. Bias can arise from any of the error sources known to affect single beam systems in addition to both translation and angle misregistration between the two sample volumes; standard deviation is the result of random temporal fluctuations in Doppler frequency estimates in each beam. Approximate closed-form expressions for both biases and standard deviations of the velocity estimates are derived, and the performance of a typical practical dual beam system is discussed as an illustration of the theory.

Velocity fluctuation reduction in vector Doppler ultrasound using a hybrid single/dual-beam algorithm.

In order to reduce the fluctuations in the velocity magnitude estimate, we propose a modification to the standard algorithm for reconstructing the (two component) vector velocity from the measured Doppler shifts in two directions. This uses the standard dual-beam algorithm, combined with temporal smoothing, to find only the velocity angle, then uses the single-beam algorithm to estimate the velocity magnitude. We present initial data showing the significant reduction in velocity estimate fluctuation that this hybrid method achieves compared to the standard algorithm.

Analysis of skin line pattern for lesion classification.

BACKGROUND/PURPOSE: It has been observed that skin patterning tends to be disrupted by malignant but not by benign skin lesions. This suggests that measurements of skin pattern disruption on simply captured white light optical skin images could be a useful contribution to a diagnostic feature set. Previous work using a measurement of line strength by a consistent high-value profiling technique followed by local variance measurement or a region agglomerative classifier to measure skin line pattern disruption was extremely promising but computationally intensive, suggesting that the idea of measuring skin pattern disruption was useful but a simpler method was required. METHODS: The skin pattern was extracted by high-pass filtration and enhanced by adaptive anisotropic (spatial variant) filtering which smoothes along skin lines but not across them. The skin line main direction and direction variance were estimated using a local image gradient matrix and the difference of these measures across the lesion image boundary was used as a lesion classifier. RESULTS: A set of images of malignant melanoma and benign naevi were processed as above and the scatter plot of results in a two-dimensional feature (line direction and line variation difference) space showed excellent separation of benign and malignant lesions. An ROC plot enclosed an area of 0.88. CONCLUSIONS: The experimental results showed that the local line direction and the local line variation were promising features for distinguishing malignant melanoma from benign lesion and the methods used were effective and computationally low-cost.