AI-Based Video Clipping of Soccer Events

The current gold standard for extracting highlight clips from soccer games is the use of manual annotations and clippings, where human operators define the start and end of an event and trim away the unwanted scenes. This is a tedious, time-consuming, and expensive task, to the extent of being rendered infeasible for use in lower league games. In this paper, we aim to automate the process of highlight generation using logo transition detection, scene boundary detection, and optional scene removal. We experiment with various approaches, using different neural network architectures on different datasets, and present two models that automatically find the appropriate time interval for extracting goal events. These models are evaluated both quantitatively and qualitatively, and the results show that we can detect logo and scene transitions with high accuracy and generate highlight clips that are highly acceptable for viewers. We conclude that there is considerable potential in automating the overall soccer video clipping process.

Synchrotron CT of an equine digit at the Australian Synchrotron Imaging and Medical Beamline

Laminitis is an extremely painful and debilitating condition of horses that can affect their athletic ability and even quality of life. The current gold standard for assessment of laminar tissue is histology, which is the only modality that enables detailed visualization of the lamina. Histology requires dissection of the hoof and therefore can only represent one specific time point. The superior spatial and contrast resolution of synchrotron computed tomography (sCT), when compared with readily available imaging modalities, such as radiographs and conventional CT, provides an opportunity for detailed studies of the lamina without the need for hoof dissection and histological assessment. If the resolution of histology can be matched or even approached, dynamic events, such as laminar blood flow, could also be studied on the microscopic tissue level. To investigate this possible application of sCT further, two objectives are presented: (i) to develop a protocol for sCT of an equine digit using cadaver limbs and (ii) to apply the imaging protocol established during (i) for sCT imaging of the vasculature within the foot using an ex vivo perfusion system to deliver the vascular contrast. The hypotheses were that sCT would allow sufficient resolution for detailed visualization to the level of the secondary lamellae and associated capillaries within the equine digit. Synchrotron CT enabled good visualization of the primary lamellae (average length 3.6 mm) and the ex vivo perfusion system was able to deliver vascular contrast agent to the vessels of the lamina. The individual secondary lamellae (average length 0.142 mm) could not be seen in detail, although differentiation between primary and secondary lamellae was achieved. This approaches, but does not yet reach, the current gold standard, histology, for assessment of the lamellae; however, with further refinement of this imaging technique, improved resolution may be accomplished in future studies.

OnePetri: accelerating common bacteriophage Petri dish assays with computer vision

IntroductionBacteriophage plaque enumeration is a critical step in a wide array of protocols. The current gold standard for plaque enumeration on Petri dishes is through manual counting. This approach is time-intensive, has low-throughput, is limited to Petri dishes which have a countable number of plaques, and can have variable results upon recount due to human error.MethodsWe present OnePetri, a collection of trained machine learning models and open-source mobile application for the rapid enumeration of bacteriophage plaques on circular Petri dishes.ResultsWhen compared against the current gold standard of manual counting, OnePetri was significantly faster, with minimal error. Compared against two other similar tools, Plaque Size Tool and CFU.AI, OnePetri had higher plaque recall and reduced detection times on most test images.ConclusionsThe OnePetri application can rapidly enumerate phage plaques on circular Petri dishes with high precision and recall.

Validation of a direct-to-PCR COVID-19 detection protocol utilizing mechanical homogenization: A model for reducing resources needed for accurate testing

Efficient and effective viral detection methodologies are a critical piece in the global response to COVID-19, with PCR-based nasopharyngeal and oropharyngeal swab testing serving as the current gold standard. With over 100 million confirmed cases globally, the supply chains supporting these PCR testing efforts are under a tremendous amount of stress, driving the need for innovative and accurate diagnostic solutions. Herein, the utility of a direct-to-PCR method of SARS-CoV-2 detection grounded in mechanical homogenization is examined for reducing resources needed for testing while maintaining a comparable sensitivity to the current gold standard workflow of nasopharyngeal and oropharyngeal swab testing. In a head-to-head comparison of 30 patient samples, this initial clinical validation study of the proposed homogenization-based workflow demonstrated significant agreeability with the current extraction-based method utilized while cutting the total resources needed in half.

Smartphone GPS signatures of patients undergoing spine surgery correlate with mobility and current gold standard outcome measures

OBJECTIVE Patient-reported outcome measures (PROMs) are currently the gold standard to evaluate patient physical performance and ability to recover after spine surgery. However, PROMs have significant limitations due to the qualitative and subjective nature of the information reported as well as the impossibility of using this method in a continuous manner. The smartphone global positioning system (GPS) can be used to provide continuous, quantitative, and objective information on patient mobility. The aim of this study was to use daily mobility features derived from the smartphone GPS to characterize the perioperative period of patients undergoing spine surgery and to compare these objective measurements to PROMs, the current gold standard. METHODS Eight daily mobility features were derived from smartphone GPS data in a population of 39 patients undergoing spine surgery for a period of 2 months starting 3weeks before surgery. In parallel, three different PROMs for pain (visual analog scale [VAS]), disability (Oswestry Disability Index [ODI]) and functional status (Patient-Reported Outcomes Measurement Information System [PROMIS]) were serially measured. Segmented linear regression analysis was used to assess trends before and after surgery. The Student paired t-test was used to compare pre- and postoperative PROM scores. Pearson’s correlation was calculated between the daily average of each GPS-based mobility feature and the daily average of each PROM score during the recovery period. RESULTS Smartphone GPS features provided data documenting a reduction in mobility during the immediate postoperative period, followed by a progressive and steady increase with a return to baseline mobility values 1 month after surgery. PROMs measuring pain, physical performance, and disability were significantly different 1 month after surgery compared to the 2 immediate preoperative weeks. The GPS-based features presented moderate to strong linear correlation with pain VAS and PROMIS physical score during the recovery period (Pearson r > 0.7), whereas the ODI and PROMIS mental scores presented a weak correlation (Pearson r approximately 0.4). CONCLUSIONS Smartphone-derived GPS features were shown to accurately characterize perioperative mobility trends in patients undergoing surgery for spine-related diseases. Features related to time (rather than distance) were better at describing patient physical and performance status. Smartphone GPS has the potential to be used for the development of accurate, noninvasive and personalized tools for patient mobility monitoring after surgery.

Pressure Relieving Support Surfaces: a Randomised Evaluation 2 (PRESSURE 2): using photography for blinded central endpoint review

Background PRESSURE 2 is a randomised evaluation of the clinical and cost-effectiveness of two types of mattress for the prevention of pressure ulcers (PUs). The primary clinical endpoint was time to development of a category ≥2 PU. The current ‘gold standard’ for PU identification is expert clinical assessment. Due to the mattress appearance, a blinded assessment of the endpoint is not possible. This poses a risk to the internal validity of the study. A possible approach is to use photographs of skin sites, with central blinded review. However, there are practical and scientific concerns including patients’ consent to photographs, burden of data collection, photograph quality, data completeness and comparison of photographs to the current ‘gold standard’. This paper reports the findings of the PRESSURE 2 photographic validation sub-study. Method Where consent was obtained, photographs were taken of all category ≥2 PUs on the first presentation to assess over-reporting, and for the assessment of under-reporting, a random sample of 10% patients had an assessment by an independent clinical assessor who also photographed two skin sites. The staff were trained in taking and transferring photographs using standardised procedures and equipment. A card included in the photograph recorded participant details and a ‘greyscale’ for correction of white balance during processing. Three blinded reviewers assessed the photographs and rated how confident they were in their assessment. Results The trial recruited 2029 patients; 85% consented to photography, and 532 photographs were received and used in the blinded central review. The level of confidence varied by skin classification with more confidence observed when the skin was assessed as being less severe than a category ≥2 PU. Overall, there was a very good reliability compared to the gold standard expert clinical assessment (87.8%, kappa 0.82). Conclusion Study findings have usefully informed the scientific and practical issues of blinded assessment of PU status to reducing the risk of bias in medical device trials. The reliability of central blinded expert photography was found to be ‘very good’ (PABAK). Photographs have been found to be an acceptable method of data validation for participants. Methods to improve the quality of photographs would increase the confidence in the assessments. Trial registration ISRCTN Registry ISRCTN01151335. Registered on 19 April 2013

Harnessing recombinase polymerase amplification for rapid detection of SARS-CoV-2 in resource-limited settings

The COVID-19 pandemic has challenged testing capacity worldwide. The mass testing needed to stop the spread of the virus requires new molecular diagnostic tests that are faster and with reduced equipment requirement, but as sensitive as the current gold standard protocols based on polymerase chain reaction.We developed a fast (25-35 minutes) molecular test using reverse transcription recombinase polymerase amplification for simultaneous detection of two conserved regions of the virus, targeting the E and RdRP genes. The diagnostic platform offers two complementary detection methods: real-time fluorescence or visual dipstick.The analytical sensitivity of the test by real-time fluorescence was 9.5 (95% CI: 7.0-18) RNA copies per reaction for the E gene and 17 (95% CI: 11-93) RNA copies per reaction for the RdRP gene. The analytical sensitivity for the dipstick readout was 130 (95% CI: 82-500) RNA copies per reaction. The assay showed high specificity with both detection methods when tested against common seasonal coronaviruses, SARS-CoV and MERS-CoV model samples. The dipstick readout demonstrated potential for point-of-care testing, with simple or equipment-free incubation methods and a user-friendly prototype smartphone application was proposed with data capture and connectivity.This ultrasensitive molecular test offers valuable advantages with a swift time-to-result and it requires minimal laboratory equipment compared to current gold standard assays. These features render this diagnostic platform more suitable for decentralised molecular testing.

FimH as a scaffold for regulated molecular recognition

Background Recognition proteins are critical in many biotechnology applications and would be even more useful if their binding could be regulated. The current gold standard for recognition molecules, antibodies, lacks convenient regulation. Alternative scaffolds can be used to build recognition proteins with new functionalities, including regulated recognition molecules. Here we test the use of the bacterial adhesin FimH as a scaffold for regulated molecular recognition. FimH binds to its native small molecule target mannose in a conformation-dependent manner that can be regulated by two types of noncompetitive regulation: allosteric and parasteric. Results We demonstrate that conformational regulation of FimH can be maintained even after reengineering the binding site to recognize the non-mannosylated targets nickel or Penta-His antibody, resulting in an up to 7-fold difference in KD between the two conformations. Moreover, both the allosteric and parasteric regulatory mechanisms native to FimH can be used to regulate binding to its new target. In one mutant, addition of the native ligand mannose parasterically improves the mutant’s affinity for Penta-His 4-fold, even as their epitopes overlap. In another mutant, the allosteric antibody mab21 reduces the mutant’s affinity for Penta-His 7-fold. The advantage of noncompetitive regulation is further illustrated by the ability of this allosteric regulator to induce 98% detachment of Penta-His, even with modest differences in affinity. Conclusions This illustrates the potential of FimH, with its deeply studied conformation-dependent binding, as a scaffold for conformationally regulated binding via multiple mechanisms.

PRESERFLO MicroShunt

The PRESERFLO® MicroShunt (Santen Pharmaceutical Co. Ltd., Osaka, Japan), formerly called the InnFocus MicroShunt®, is a trans-scleral device that shunts aqueous humour from the anterior chamber to a filtering bleb under the conjunctiva and Tenon’s capsule. Manufactured from an inert biocompatible material called poly(styrene-block-isobutylene-block-styrene), or ‘SIBS’, the device elicits minimal foreign body reaction and inflammation; potentially reducing the risk of bleb-related fibrosis and failure. The MicroShunt is 8.5 mm long with a 70 μm lumen and is designed to minimize hypotony based on the Hagen–Poiseuille equation. Inserted via an ab-externo approach, the MicroShunt eliminates the need for creation of a scleral flap, sclerostomy, iridectomy, scleral flap suturing and postoperative suture lysis. Clinical trials show promising results with the MicroShunt achieving intraocular pressure reduction approaching that of trabeculectomy, the current gold standard for treating refractory glaucoma.