Search

Medical image segmentation is a crucial element of computer-aided diagnosis (CAD) systems. Segmentation maps are used to calculate imaging features, such as quantitative disease distribution and radiomic features. Since their introduction in 2015, UNets have become the state-of-the-art segmentation tools. However, since that time, many new methods for image processing have been introduced, such as vision transformers and multi-layer-perceptron-mixers (MLP-Mixers). Alongside baseline UNets, we have now investigated the application of such MLP-Mixers for medical image segmentation, as part of a CAD system for the diagnosis of interstitial lung diseases (ILDs). Furthermore, we have investigated the effect of 2D and 3D data representations on segmentation and the final CAD results. We have evaluated the performance of the baseline segmentation methods and the MLP-Mixer primary on the overall diagnostic performance of the CAD system - as well as on the accuracy of segmentation as an intermediate step. In addition to network and data representation variations, we have investigated two different techniques for selecting features, an agnostic method and an alternative approach which selects features tailored to a specific segmentation map and diagnosis task. Finally, the CAD’s performance was compared with that of four independent specialists in chest radiology. Among the 105 test cases, the diagnostic accuracy was 77.2±1.6% for the AI-approaches and 79.0±6.9% for the radiologists, indicating that the proposed systems perform comparably well to human readers in most of the cases. For the task of ILD pattern segmentation, similar results were obtained with 3D data and 2D tomography slices.

Abstract Background Triggering final oocyte maturation is a pivotal step in modern patient-tailored IVF/ICSI treatment, securing the optimal number of mature oocytes retrieved without compromising fertilization, embryo development, and live birth. Several factors need to be considered when deciding the time of the trigger: the size of the leading follicles, distribution of the follicular cohort, the duration of stimulation, the protocol used for stimulation, and ovarian response status. Main body The current narrative review aims to appraise all available evidence for determining the proper time for inducing final oocyte maturation following IVF treatment. Moreover, it discusses the impact of the stimulation protocol, follicular size, and magnitude of ovarian response on choosing the proper timing for trigger. Comprehensive literature search of all available articles and relevant articles studying the criteria for timing of final oocyte maturation trigger in IVF/ICSI cycles were included in this review. It was found that leading follicles size of 16–22 mm is associated with the optimum oocyte maturation ratio, size of the remaining cohort of follicles should be ≥ 14 mm, 10–12 days of minimum length of stimulation should be auspicated in normal responders before trigger, and the timing of trigger administration should not depend solely on hormonal levels. Conclusion In conclusion, the timing of triggering of final oocyte maturation in ICSI cycles should be individualized on a case-by-case basis.

Purpose: Sperm DNA fragmentation (SDF) has been associated with male infertility and poor outcomes of assisted reproductive technology (ART). The purpose of this study was to investigate global practices related to the management of elevated SDF in infertile men, summarize the relevant professional society recommendations, and provide expert recommendations for managing this condition. Materials and Methods: An online global survey on clinical practices related to SDF was disseminated to reproductive clinicians, according to the CHERRIES checklist criteria. Management protocols for various conditions associated with SDF were captured and compared to the relevant recommendations in professional society guidelines and the appropriate available evidence. Expert recommendations and consensus on the management of infertile men with elevated SDF were then formulated and adapted using the Delphi method.Results: A total of 436 experts from 55 different countries submitted responses. As an initial approach, 79.1% of reproductive experts recommend lifestyle modifications for infertile men with elevated SDF, and 76.9% prescribe empiric antioxidants. Regarding antioxidant duration, 39.3% recommend 4–6 months and 38.1% recommend 3 months. For men with unexplained or idiopathic infertility, and couples experiencing recurrent miscarriages associated with elevated SDF, most respondents refer to ART 6 months after failure of conservative and empiric medical management. Infertile men with clinical varicocele, normal conventional semen parameters, and elevated SDF are offered varicocele repair immediately after diagnosis by 31.4%, and after failure of antioxidants and conservative measures by 40.9%. Sperm selection techniques and testicular sperm extraction are also management options for couples undergoing ART. For most questions, heterogenous practices were demonstrated. Conclusions: This paper presents the results of a large global survey on the management of infertile men with elevated SDF and reveals a lack of consensus among clinicians. Furthermore, it demonstrates the scarcity of professional society guidelines in this regard and attempts to highlight the relevant evidence. Expert recommendations are proposed to help guide clinicians.

Abstract Research question Does a frozen-embryo transfer in an artificially-prepared endometrium (FET-HRT) cycle yield similar clinical pregnancy rate with 7 days of oestrogen priming compared to 14 days? Design This is a single-centre, randomized, controlled, open-label pilot study. All FET-HRT cycles were performed in a tertiary centre between October 2018 and January 2021. Overall, 160 patients were randomized, with a 1:1 allocation, into two groups of 80 patients each: group A (7 days of E2 prior to P4 supplementation) and group B (14 days of E2 prior to P4 supplementation). Both groups received single blastocyst stage embryos on the 6th day of vaginal P4 administration. The primary outcome was the feasibility of such strategy assessed as clinical pregnancy rate, secondary outcomes were biochemical pregnancy rate, miscarriage rate, live birth rate and serum hormone levels on the day of FET. Chemical pregnancy was assessed by an hCG blood test 12 days after FET and clinical pregnancy was confirmed by transvaginal ultrasound at 7 weeks. Results The analysis included 160 patients who were randomly assigned to either group A or group B on the seventh day of their FET-HRT cycle if the measured endometrial thickness was above 6.5 mm. Following screening failures and of drop-outs, 144 patients were finally included both in group A (75 patients) or group B (69 patients). Demographic characteristics for both groups were comparable. The biochemical pregnancy rate was 42.5% and 48.8% for group A and group B, respectively (p 0.526). Regarding the clinical pregnancy rate at 7 weeks, no statistical difference was observed (36.3% vs 46.3% for group A and group B, respectively, p = 0.261). The secondary outcomes of the study (biochemical pregnancy, miscarriage, and live birth rate) were comparable between the two groups for IIT analysis, as well as the P4 values on the day of FET. Conclusions In a frozen embryo transfer cycle, performed with artificial preparation of the endometrium, 7 versus 14 days of oestrogen priming are comparable, in terms of clinical pregnancy rate; the advantages of a seven-day protocol include the shorter time to pregnancy, reduced exposure to oestrogens, and more flexibility of scheduling and programming, and less probability to recruit a follicle and have a spontaneous LH surge. It is important to keep in mind that this study was designed as a pilot trial with a limited study population as such it was underpowered to determine the superiority of an intervention over another; larger-scale RCTs are warranted to confirm our preliminary results. Trial registration Clinical trial number: NCT03930706.

Purpose: Sperm DNA fragmentation (SDF) testing was recently added to the sixth edition of the World Health Organization laboratory manual for the examination and processing of human semen. Many conditions and risk factors have been associated with elevated SDF; therefore, it is important to identify the population of infertile men who might benefit from this test. The purpose of this study was to investigate global practices related to indications for SDF testing, compare the relevant professional society guideline recommendations, and provide expert recommendations. Materials and Methods: Clinicians managing male infertility were invited to take part in a global online survey on SDF clinical practices. This was conducted following the CHERRIES checklist criteria. The responses were compared to professional society guideline recommendations related to SDF and the appropriate available evidence. Expert recommendations on indications for SDF testing were then formulated, and the Delphi method was used to reach consensus. Results: The survey was completed by 436 experts from 55 countries. Almost 75% of respondents test for SDF in all or some men with unexplained or idiopathic infertility, 39% order it routinely in the work-up of recurrent pregnancy loss (RPL), and 62.2% investigate SDF in smokers. While 47% of reproductive urologists test SDF to support the decision for varicocele repair surgery when conventional semen parameters are normal, significantly fewer general urologists (23%; p=0.008) do the same. Nearly 70% would assess SDF before assisted reproductive technologies (ART), either always or for certain conditions. Recurrent ART failure is a common indication for SDF testing. Very few society recommendations were found regarding SDF testing. Conclusions: This article presents the largest global survey on the indications for SDF testing in infertile men, and demonstrates diverse practices. Furthermore, it highlights the paucity of professional society guideline recommendations. Expert recommendations are proposed to help guide clinicians.

BackgroundScreening programs that pre-emptively and routinely test population groups for disease at a massive scale were first implemented during the COVID-19 pandemic in a handful of countries. One of these countries was Greece, which implemented a mass self-testing program during 2021. In contrast to most other non-pharmaceutical interventions (NPIs), mass self-testing programs are particularly attractive for their relatively small financial and social burden, and it is therefore important to understand their effectiveness to inform policy makers and public health officials responding to future pandemics. This study aimed to estimate the number of deaths and hospitalizations averted by the program implemented in Greece and evaluate the impact of several operational decisions.MethodsGranular data from the mass self-testing program deployed by the Greek government between April and December 2021 were obtained. The data were used to fit a novel compartmental model that was developed to describe the dynamics of the COVID-19 pandemic in Greece in the presence of self-testing. The fitted model provided estimates on the effectiveness of the program in averting deaths and hospitalizations. Sensitivity analyses were used to evaluate the impact of operational decisions, including the scale of the program, targeting of sub-populations, and sensitivity (i.e., true positive rate) of tests.ResultsConservative estimates show that the program reduced the reproduction number by 4%, hospitalizations by 25%, and deaths by 20%, translating into approximately 20,000 averted hospitalizations and 2,000 averted deaths in Greece between April and December 2021.ConclusionMass self-testing programs are efficient NPIs with minimal social and financial burden; therefore, they are invaluable tools to be considered in pandemic preparedness and response.

Time series of financial data are both frequent and important in everyday practice. Numerous applications are based, for example, on time series of asset prices or market indices. In this article, the application of fractal interpolation functions in modelling financial time series is examined. Our motivation stems from the fact that financial time series often present fluctuations or abrupt changes which the fractal interpolants can inherently model. The results indicate that the use of fractal interpolation in financial applications is promising.

This paper compares three finite element-based methods used in a physics-based non-rigid registration approach and reports on the progress made over the last 15 years. Large brain shifts caused by brain tumor removal affect registration accuracy by creating point and element outliers. A combination of approximation- and geometry-based point and element outlier rejection improves the rigid registration error by 2.5 mm and meets the real-time constraints (4 min). In addition, the paper raises several questions and presents two open problems for the robust estimation and improvement of registration error in the presence of outliers due to sparse, noisy, and incomplete data. It concludes with preliminary results on leveraging Quantum Computing, a promising new technology for computationally intensive problems like Feature Detection and Block Matching in addition to finite element solver; all three account for 75% of computing time in deformable registration.

In the modern era, patients often resort to the internet for answers to their health-related concerns, and clinics face challenges to providing timely response to patient concerns. This has led to a need to investigate the capabilities of AI chatbots for ophthalmic diagnosis and triage. In this in silico study, 80 simulated patient complaints in ophthalmology with varying urgency levels and clinical descriptors were entered into both ChatGPT and Bard in a systematic 3-step submission process asking chatbots to triage, diagnose, and evaluate urgency. Three ophthalmologists graded chatbot responses. Chatbots were significantly better at ophthalmic triage than diagnosis (90.0% appropriate triage vs. 48.8% correct leading diagnosis; p < 0.001), and GPT-4 performed better than Bard for appropriate triage recommendations (96.3% vs. 83.8%; p = 0.008), grader satisfaction for patient use (81.3% vs. 55.0%; p < 0.001), and lower potential harm rates (6.3% vs. 20.0%; p = 0.010). More descriptors improved the accuracy of diagnosis for both GPT-4 and Bard. These results indicate that chatbots may not need to recognize the correct diagnosis to provide appropriate ophthalmic triage, and there is a potential utility of these tools in aiding patients or triage staff; however, they are not a replacement for professional ophthalmic evaluation or advice.

The DIAD beamline for Dual Imaging and Diffraction at Diamond Light Source has opted to use an industrial robot to position its Dectris Pilatus 2M CdTe diffraction detector. This setup was chosen to enable flexible positioning of the detector in a quarter-sphere around the sample position whilst reliably holding the large weight of 139 kg of detector, detector mount and cabling in a stable position. Metrology measurements showed that the detector can be positioned with a linear repeatability of

Background/aims Retinal microvascular ischaemia may produce localised middle retinal disruption with corresponding scotoma, a phenomenon termed paracentral acute middle maculopathy (PAMM). Small chronic middle retinal atrophic lesions termed retinal ischaemic perivascular lesions (RIPLs) appear qualitatively similar to PAMM lesions and have recently been hypothesised to result specifically from PAMM. However, no studies have quantitatively demonstrated an ischaemic origin of RIPLs. We quantitatively investigated the pathophysiology of RIPLs and their relationship with PAMM with swept-source optical coherence tomography angiography (SS-OCTA).Methods A total of 14 controls and 25 patients being evaluated for carotid artery stenosis (CAS) were enrolled. SS-OCTA imaging of each eye was taken. Projection-resolved en face 6 mm × 6 mm superficial capillary plexus (SCP) and deep capillary plexus (DCP) images were quantitatively analysed with two algorithms for changes in vessel linear density (VLD) and vessel tortuosity (VT) at RIPLs relative to both the immediately surrounding macula and the entire macula, as well as between eyes with RIPLs and eyes without RIPLs.Results All controls and 22 of 25 CAS patients were included in the analysis. RIPLs demonstrated a localised decrease in DCP VLD in CAS patients and controls. RIPLs tended to show a localised decrease in SCP VLD in CAS patients but a localised increase in controls. No changes in VT were found. Eyes with RIPLs had VLD and VT similar to their RIPL-free fellow eyes.Conclusion RIPLs are associated with quantifiable local, but not global, ischaemia, supporting the idea of shared pathophysiology with classic PAMM lesions along a continuum of ischaemia severity.

Listeners with hearing loss often struggle to understand speech in noise, even with a hearing aid. To better understand the auditory processing deficits that underlie this problem, we made large-scale brain recordings from gerbils, a common animal model for human hearing, while presenting a large database of speech and noise sounds. We first used manifold learning to identify the neural subspace in which speech is encoded and found that it is low-dimensional and that the dynamics within it are profoundly distorted by hearing loss. We then trained a deep neural network (DNN) to replicate the neural coding of speech with and without hearing loss and analyzed the underlying network dynamics. We found that hearing loss primarily impacts spectral processing, creating nonlinear distortions in cross-frequency interactions that result in a hypersensitivity to background noise that persists even after amplification with a hearing aid. Our results identify a new focus for efforts to design improved hearing aids and demonstrate the power of DNNs as a tool for the study of central brain structures.

Oriented polymers may exhibit high thermal conductivity, but the upper limits remain unclear. Here, the authors show that heat-carrying phonons scatter primarily from crystalline domain boundaries in disentangled polyethylene films.

Drakopoulos et al developed a machine-learning and computational-neuroscience approach that transforms analytical models of sensory neurons and synapses into deep-neural-network (DNN) neuronal units with the same biophysical properties. Focusing on auditory neurons and synapses, they showed that their DNN-model architecture could be extended to a variety of existing analytical models and to other neuron and synapse types, thus potentially assisting the development of large-scale brain networks and DNN-based treatments.

Entanglement dynamics of uniaxially-oriented, disentangled ultra-high molecular weight polyethylene nanocomposites modified with gold nanoparticles are investigated, using dielectric spectroscopy, during the transition to melt state. The dc conduction is approximately calculated via the logarithmic derivative of dielectric permittivity and is observed to decrease with the formation of entanglements. As the draw ratio increases, the progressive formation of entanglements resulted in a stronger dc conductivity decrease due to the loss of orientation in the pre-melt crystalline and partially oriented amorphous chain segments. Additionally, a sharp peak is obtained in the absence of dc conductivity, attributed to the dipolar contribution of the Maxwell-Wagner-Sillars (MWS) interfacial polarization between the gold nanoparticles and the polymer chains. The relaxation time of the MWS interfacial polarization increases with the progressive formation of entanglements, as observed in a previous study. The results presented shed light on the process of entanglement formation in oriented ultrahigh molecular weight polyethylene nanocomposites.

The scope of this article is to identify the parameters of bivariate fractal interpolation surfaces by using convex hulls as bounding volumes of appropriately chosen data points so that the resulting fractal (graph of) function provides a closer fit, with respect to some metric, to the original data points. In this way, when the parameters are appropriately chosen, one can approximate the shape of every rough surface. To achieve this, we first find the convex hull of each subset of data points in every subdomain of the original lattice, calculate the volume of each convex polyhedron and find the pairwise intersections between two convex polyhedra, i.e., the convex hull of the subdomain and the transformed one within this subdomain. Then, based on the proposed methodology for parameter identification, we minimise the symmetric difference between bounding volumes of an appropriately selected set of points. A methodology for constructing continuous fractal interpolation surfaces by using iterated function systems is also presented.