Interpretation Of Imaging Data Research Articles

Integrating Widespread Coral Reef Monitoring Tools for Managing both Area and Point Annotations

Abstract. Large-area image acquisition techniques are essential in underwater investigations: high-resolution 3D image-based reconstructions have improved coral reef monitoring by enabling novel seascape ecological analysis. Artificial intelligence (AI) offers methods for significantly accelerating image data interpretation, such as automatically recognizing, enumerating, and measuring organisms. However, the rapid proliferation of these technological achievements has led to a relative lack of standardization of methods. Remarkably, there are notable differences in procedures for generating human and AI annotations, and there is also a scarcity of publicly available datasets and shared machine-learning models. The lack of standard procedures makes it challenging to compare and reproduce scientific findings. One way to overcome this problem is to make the most used platforms by coral reef scientists interoperable so that the analyses can all be exported into a common format. This paper introduces functionality to promote interoperability between three popular open-source software tools dedicated to the digital study of coral reefs: TagLab, CoralNet, and Viscore. As users of each platform may have different analysis pipelines, we discuss several workflows for managing and processing point and area annotations, improving collaboration among these tools. Our work sets the foundation for a more seamless ecosystem that maintains the established investigation procedures of various laboratories but allows for easier result sharing.

Do Cone Beam CT Picture Archiving and Communication Systems Viewer Interfaces Meet the Expectations of Dental Professionals From a Usability Perspective?

Background Cone beam computed tomography (CBCT) has revolutionized dental and maxillofacial imaging by providing high-resolution 3D visualizations, essential for accurate diagnosis and treatment planning. Despite its clinical advancements, the usability of CBCT viewer interfaces, which play a crucial role in the effective interpretation of imaging data, remains a critical concern. Objective This study aims to evaluate the usability of a CBCT viewer interfaceinclinical settings, focusing on theeffectiveness, efficiency, and satisfaction perspectives, to identify potential areas for improvement. Methods Twenty-two participants (N=22) were assigned the task of locating the mental foramen in a mandible dataset, selected randomly, using the multiplanar reconstruction (MPR) mode of a CBCT viewer interface on a local network. The task's effectiveness was gauged by the completion rate, while efficiency was measured through the duration of the task, the number of mouse clicks, and the cursor's path in both pixels and meters. Satisfaction or perceived usability was evaluated using the system usability scale (SUS-TR), and computer system usability questionnaire (T-CSUQ), among other scales, with participants also providing open-ended feedback. Demographic characteristics served as classification variables. Results All participants completed the given task. No demography-related differences in efficiency were observed. Perceived usability (SUS-TR: 60.68±19.58, T-CSUQ: 43.63±16.34) was below the industry standards, categorizing the system as a detractor. Commonly reported issues included accessing the MPR window, cursor behavior, and unclear error messages. The mean SUS-TR score negatively correlated with efficiency-related variables (p<0.05 for each). Conclusions The study's findings indicate that the CBCT viewer interface does not fully meet dental professionals' usability expectations, as evidenced by the task's completion rate, efficiency metrics, and below-average usability scores. Despite the successful task completion by all participants, the identified issues in interface design, such as difficulties in accessing the MPR window and unclear error messages, highlight significant areas for improvement. To enhance user satisfaction and efficiency, future developments of CBCT viewer interfaces should incorporate feedback from end-users and prioritize user-friendly design principles.

Analyzing in situ stresses and wellbore stability in one of the south Iranian hydrocarbon gas reservoirs

This study aims to analyze in situ stresses and wellbore stability in one of the Iranian gas reservoirs by using well log data, including density, sonic (compressional and shear slowness), porosity, formation micro-image (FMI) logs, modular formation dynamics tester (MDT), and rock mechanical tests. The high burial depth, high pore pressure, and strike-slip stress regime of the field require an optimal design of geomechanical parameters based on an integrated data set consisting of static and dynamic data, which is available for this study. Firstly, poroelastic modulus and vertical stress were calculated. Afterward, the Eaton’s equation was used to estimate pore pressure from well logging data. The geomechanical parameters were also calibrated through the interpretation of image data, the use of the modular formation dynamics tester (MDT), and laboratory rock mechanic tests. Employing poroelastic equations, the lowest and highest horizontal stresses were calculated. It was shown that the maximum horizontal stress and minimum horizontal stress correspond to sigma H and sigma h, indicating the strike-slope fault regime. The findings of this research indicated that the equivalent mud weight (EMW) resulted in 10–13 ppg suitable for the Kangan Formation and 11–14 ppg suitable for the Dalan Formation. Additionally, the well azimuth in the NE-SW direction provided the best stability for drilling the encountered formations. Therefore, the results of this study serve as cost-effective tools in planning adjacent wells in carbonate formations of gas field to predict the wellbore stability and safe mud window.

Distribusi Spasial Lamun di Perairan Sekotong Barat-TWP Gita Nada Lombok Barat Menggunakan Citra Landsat 8 OLI

West Sekotong Village is one of the coastal water areas included in the Gita Nada Aquatic Tourism Park (TWP) area which has a seagrass habitat. Seagrass beds in West Sekotong are threatened due to increased tourism and community activities. This research aims to determine the distribution and extent of seagrass beds on the coast of West Sekotong-TWP Gita Nada. This research uses a remote sensing approach through interpretation of Landsat 8 OLI imagery. This research consists of 2 steps, namely image data processing and supporting data. Image data processing includes several stages, namely image data collection, geometric correction, radiometric correction, image cutting, image composite, image data sharpening, image data interpretation, accuracy testing, area measurement, and final layout. Supporting data was obtained by searching related literature, interviews and surveys of activities that have the potential to have an impact on seagrass ecosystems. The research results show that the distribution of seagrass in West Sekotong-TWP Gita Nada covers the coast of the large island (mainland) and the coast of small islands. The distribution of seagrass on the mainland is found on the coasts of Medang, Tanjung Kelor, Batu Kijuk, Tawun, Labu, Kelapa, Pandanan, Gili Genting, Labuan Petung, Temeran, and Gawah Pudak. In the coastal areas of small islands, seagrass is found on Gili Sudak, Gili Tangkong, Gili Nanggu, Gili Poh, and Gili Lontar. The total area of seagrass beds in West Sekotong is around 144.68 ha. On the mainland coast the area is 118.96 ha (82.2%) and on small islands the area is around 25.72 ha (17.8%). The distribution of seagrass in West Sekotong is influenced by many factors, among which those observed in this research are the type of substrate, the presence of river estuaries, and the use of land and coastal waters.

Peri-device leak after left atrial appendage closure: echocardiographic characteristics and thromboembolism risk

Abstract Background Structural variations in left atrial appendage (LAA) ostial morphology can contribute to incomplete closure of the LAA and peri-device leak (PDL) after LAA closure (LAAC). Prior data suggests that there is an increased residual risk of thromboembolism in patients with PDL after LAAC. However, there is a paucity of information on the interpretation of imaging data to risk stratify patients with PDL. Purpose We evaluated echocardiographic characteristics on post-LAAC imaging, and their association with residual risk of thromboembolism. Methods We analyzed data from follow up imaging with transesophageal echocardiograms (TEEs) which were performed at ∼1-year follow up after the LAAC with either the Watchman or Watchman-FLX device (Boston Scientific Inc). The images were reviewed in a blinded manner with regard to thromboembolic events after LAAC. In our analyses, we further classified the TEE images into 3 categories: 1) no leak, and completely thrombosed distal LAA, 2) no leak, but echolucent distal LAA (indicative of some residual, but undetectable, connection), or 3) presence of leak and echolucent distal LAA (Figure 1). Results A total of 650 patients underwent LAAC at our center between 3/2015 and 12/2021: age 75.8±8.9 years, 64% male, and CHA2DS2-Vasc 4.2±1.4. TEE imaging data (performed at 1-year or later after LAAC) was available for review in 460 patients. In 238 of 460 patients (51.7%) there was no leak and the LAA was completely thrombosed (Cohort 1). In 83 of 460 (18%) patients there no detectable leak but the LAA was echolucent (Cohort 2), and in 139 of 460 (30.2%) patients there was a discernible leak and the LAA was echolucent (Cohort 3). In this subset of 460 patients, there were 29 (3.2% per year) thromboembolic events (stroke or transient ischemic attack) over 2.0±0.2 years of follow-up; of these, only 3 (0.6% per year) strokes were disabling, and none were fatal. The median NIH stroke scale in patients with stroke/TIA was 2 (range 0-14; NIHSS &amp;lt; 5 is considered mild to moderate severity). During follow-up (2.0±0.2 years), the incidence of thromboembolism was highest in Cohort 3 (7.5% per year) in comparison to Cohort 2 (3.1% per year) and Cohort 1 (0.84% per year) (Figure 2). Upon analysis with 2 separate multivariable models, the risk of thromboembolism was highest in Cohort 3 vs Cohort 1 (adjusted OR 10.0, 95% CI: 3.3-30.0; p &amp;lt; 0.001) after adjusting for age, gender, CHA2DS2-VASc and LVEF. In contrast, when comparing Cohort 2 vs Cohort 1 in a multivariable model, the difference in thromboembolic risk was an adjusted OR 3.4 (95% CI: 0. 9-13.1; p = 0.07). Conclusions The presence of PDL after LAAC is not uncommon and confers a significant residual thromboembolic risk in patients with AF. TEE imaging characteristics aid in risk stratification of patients with PDL, and may help guide management decisions such as leak closure or resumption/continuation of anticoagulation to mitigate stroke risk.Figure 1:Classification based on reviewFigure 2:Comparison of incidence of th

Implementasi Metode Simple Multi Attribute Rating Technique (SMART) dalam Pemilihan Zona Prioritas dan Alternatif Berbasis Data Klasifikasi Indeks Vegetasi

Vegetation index analysis using the Normalized Difference Vegetation Index (NDVI) model needs to be processed using a decision support model to follow up on the Landsat 8/9 Operational Land Imaginer (OLI) satellite image data interpretation results. However, studies using the Simple Multi-Attribute Rating Technique (SMART) method to determine priority zones based on vegetation index classification data are still limited. This study uses the SMART decision support model to process NDVI classification data in mangrove areas. The stages in this study consist of four parts: the data collection stage, the data processing stage; the data analysis stage; and the data interpretation stage. At the data collection stage, the raster data used was sourced from the United States Geology Survey (USGS) platform, namely Landsat 8/9 OLI with coordinate raster data (Lat 01o43'18" N, Lon: 128o04'15" E) in 2013, 2018, and 2023. In addition, video and aerial photographs at the study site were taken using drones (Phantom 4 Version 2). At the data processing stage, the model used in calculating raster data is NDVI using the QGIS 3.30.1 application. This research data analysis and interpretation stage uses the SMART decision support model. The SMART decision support model is used to produce recommendations for priority zones for mangrove ecotourism development based on the results of the NDVI classification (minimum value, average value, maximum value) adjusted to the Decree of the State Minister of Environment Number 201 of 2004 concerning standard criteria and guidelines for mangrove forest destruction (rare, medium, and dense). Based on the calculation of the utility value of criterion C1 as a cost with a weight of 0.50 in the NDVI classification data for 2023, the second observation station is recommended as a priority zone with a total value of 0.50. Meanwhile, the calculation of the utility value of criterion C3 as a cost with a weight of 0.50 in the NDVI classification data in 2023 recommended the third observation station as a priority zone with a total value of 0.88. This means that the SMART method can be used to identify and analyze priority and alternative zones for the sustainable development of mangrove ecotourism areas.

Stroke-GFCN: ischemic stroke lesion prediction with a fully convolutional graph network.

The interpretation of image data plays a critical role during acute brain stroke diagnosis, and promptly defining the requirement of a surgical intervention will drastically impact the patient's outcome. However, determining stroke lesions purely from images can be a daunting task. Many studies proposed automatic segmentation methods for brain stroke lesions from medical images in different modalities, though heretofore results do not satisfy the requirements to be clinically reliable. We investigate the segmentation of brain stroke lesions using a geometric deep learning model that takes advantage of the intrinsic interconnected diffusion features in a set of multi-modal inputs consisting of computer tomography (CT) perfusion parameters. We propose a geometric deep learning model for the segmentation of ischemic stroke brain lesions that employs spline convolutions and unpooling/pooling operators on graphs to excerpt graph-structured features in a fully convolutional network architecture. In addition, we seek to understand the underlying principles governing the different components of our model. Accordingly, we structure the experiments in two parts: an evaluation of different architecture hyperparameters and a comparison with state-of-the-art methods. The ablation study shows that deeper layers obtain a higher Dice coefficient score (DCS) of up to 0.3654. Comparing different pooling and unpooling methods shows that the best performing unpooling method is the proportional approach, yet it often smooths the segmentation border. Unpooling achieves segmentation results more adapted to the lesion boundary corroborated with systematic lower values of Hausdorff distance. The model performs at the level of state-of-the-art models without optimized training methods, such as augmentation or patches, with a DCS of . We proposed and evaluated an end-to-end trainable fully convolutional graph network architecture using spline convolutional layers for the ischemic stroke lesion prediction. We propose a model that employs graph-based operations to predict acute stroke brain lesions from CT perfusion parameters. Our results prove the feasibility of using geometric deep learning to solve segmentation problems, and our model shows a better performance than other models evaluated. The proposed model achieves improved metric values for the DCS metric, ranging from 8.61% to 69.05%, compared with other models trained under the same conditions. Next, we compare different pooling and unpooling operations in relation to their segmentation results, and we show that the model can produce segmentation outputs that adapt to irregular segmentation boundaries when using simple heuristic unpooling operations.

A comparison of treating physician versus independent core lab assessments of post-aneurysm treatment imaging outcomes: an analysis of prospectively collected data from a randomized trial.

Aneurysm occlusion has been used as surrogate marker of aneurysm treatment efficacy. Aneurysm occlusion scales are used to evaluate the outcome of endovascular aneurysm treatment and to monitor recurrence. These scales, however, require subjective interpretation of imaging data, which can reduce the utility and reliability of these scales and the validity of clinical studies regarding aneurysm occlusion rates. Use of a core lab with independent blinded reviewers has been implemented to enhance the validity of occlusion rate assessments in clinical trials. The degree of agreement between core labs and treating physicians has not been well studied with prospectively collected data. In this study, the authors analyzed data from the Hydrogel Endovascular Aneurysm Treatment (HEAT) trial to assess the interrater agreement between the treating physician and the blinded core lab. The HEAT trial included 600 patients across 46 sites with intracranial aneurysms treated with coiling. The treating site and the core lab independently reviewed immediate postoperative and follow-up imaging (3-12 and 18-24 months, respectively) using the Raymond-Roy occlusion classification (RROC) scale, Meyer scale, and recanalization survey. A post hoc analysis was performed to calculate interrater reliability using Cohen's kappa. Further analysis was performed to assess whether degree of agreement varied on the basis of various factors, including scale used, timing of imaging, size of the aneurysm, imaging modality, location of the aneurysm, dome-to-neck ratio, and rupture status. Minimal interrater agreement was noted between the core lab reviewers and the treating physicians for assessing aneurysm occlusion using the RROC grading scale (k = 0.39, 95% CI 0.38-0.40) and Meyer scale (k = 0.23, 95% CI 0.14-0.38). The degree of agreement between groups was slightly better but still weak for assessing recanalization (k = 0.45, 95% CI 0.38-0.52). Factors that significantly improved degree of agreement were scales with fewer variables, greater time to follow-up, imaging modality (digital subtraction angiography), and wide-neck aneurysms. Assessment of aneurysm treatment outcome with commonly used aneurysm occlusion scales suffers from risk of poor interrater agreement. This supports the use of independent core labs for validation of outcome data to minimize reporting bias. Use of outcome tools with fewer point categories is likely to provide better interrater reliability. Therefore, the outcome assessment tools are ideal for clinical outcome assessment provided that they are sensitive enough to detect a clinically significant change.

Axonemal regulation by curvature explains sperm flagellar waveform modulation.

Flagellar motility is critical to natural and many forms of assisted reproduction. Rhythmic beating and wave propagation by the flagellum propels sperm through fluid and enables modulation between penetrative progressive motion, activated side-to-side yaw and hyperactivated motility associated with detachment from epithelial binding. These motility changes occur in response to the properties of the surrounding fluid environment, biochemical activation state, and physiological ligands, however, a parsimonious mechanistic explanation of flagellar beat generation that can explain motility modulation is lacking. In this paper, we present the Axonemal Regulation of Curvature, Hysteretic model, a curvature control-type theory based on switching of active moment by local curvature, embedded within a geometrically nonlinear elastic model of the flagellum exhibiting planar flagellar beats, together with nonlocal viscous fluid dynamics. The biophysical system is parameterized completely by four dimensionless parameter groupings. The effect of parameter variation is explored through computational simulation, revealing beat patterns that are qualitatively representative of penetrative (straight progressive), activated (highly yawing) and hyperactivated (nonprogressive) modes. Analysis of the flagellar limit cycles and associated swimming velocity reveals a cusp catastrophe between progressive and nonprogressive modes, and hysteresis in the response to changes in critical curvature parameter. Quantitative comparison to experimental data on human sperm exhibiting typical penetrative, activated and hyperactivated beats shows a good fit to the time-average absolute curvature profile along the flagellum, providing evidence that the model is capable of providing a framework for quantitative interpretation of imaging data.

What should AI see? Using the public’s opinion to determine the perception of an AI

Deep neural networks (DNN) have made impressive progress in the interpretation of image data so that it is conceivable and to some degree realistic to use them in safety critical applications like automated driving. From an ethical standpoint, the AI algorithm should take into account the vulnerability of objects or subjects on the street that ranges from “not at all”, e.g. the road itself, to “high vulnerability” of pedestrians. One way to take this into account is to define the cost of confusion of one semantic category with another and use cost-based decision rules for the interpretation of probabilities, which are the output of DNNs. However, it is an open problem how to define the cost structure, who should be in charge to do that, and thereby define what AI-algorithms will actually “see”. As one possible answer, we follow a participatory approach and set up an online survey to ask the public to define the cost structure. We present the survey design and the data acquired along with an evaluation that also distinguishes between perspective (car passenger vs. external traffic participant) and gender. Using simulation based F-tests, we find highly significant differences between the groups. These differences have consequences on the reliable detection of pedestrians in a safety critical distance to the self-driving car. We discuss the ethical problems that are related to this approach and also discuss the problems emerging from human–machine interaction through the survey from a psychological point of view. Finally, we include comments from industry leaders in the field of AI safety on the applicability of survey based elements in the design of AI functionalities in automated driving.

Inpatient COVID-19 vaccination rollout: Improving access to vaccination

We were aware of high numbers of inpatients unvaccinated against COVID-19 at Guy's and St Thomas' NHS Foundation Trust (GSTT). Due to this, an inpatient vaccination protocol was set up in July 2021, with initially limited uptake. From October 2021, a multidisciplinary team worked to improve the protocol for inpatient vaccination, with the development of a system that gave ownership to clinical teams. In 4 months (July 2021 to November 2021), 20 inpatients had been vaccinated at GSTT. Following our intervention, rates of uptake increased, and 34 patients were vaccinated in less than 2 months (November 2021 to January 2022). Forty-five patients who had been referred were discharged without vaccination; attempts were made to invite them to receive a vaccine. An improved pathway and referral process increased the number of inpatient vaccinations delivered. Further work is required in order to ensure that more patients who have been referred are vaccinated.

Comparing the accuracy between shear wave elastography and strain elastography in the diagnosis of breast tumors: Systematic review and meta-analysis.

Background:Shear wave elastography and strain elastography are two new ultrasonic techniques developed rapidly in recent years. Changes in tissue elasticity occur after normal tissue changes. Elastography technique transforms the elastic information of tissue into optical information for display. Thus more intuitive display of tissue elasticity. Due to the differences in principles and related imaging parameters between the two elastic imaging methods, and the acquisition and interpretation of image data in strain elastic imaging method largely depends on the experience of inspectors, and due to the significant differences between the techniques of inspectors, As a result, conflicting results have been obtained in different scholars’ studies on the accuracy comparison of the two elastography techniques in the diagnosis of breast tumors. This meta-analysis aims to compare the accuracy of the two elastography methods in the diagnosis of breast tumors, so as to provide more accurate diagnostic means for patients with breast tumors. The final results will show which elastography method is more accurate in the diagnosis of breast tumors, reduce unnecessary biopsies and provide a reference for clinical decision making.Methods:We will examine published and unpublished randomized controlled trials, observational studies and abstracts without publication type or language restrictions, and search relevant literatures in PubMed, Web of Science, Wanfang Database, CNQI and other databases until December 30, 2020. The authors will independently search relevant literature records, scan titles and abstracts, full text, collect data and assess the risk of bias. Data will be analyzed by using Meta Disc1.4 software and Stata14.0 software. Heterogeneity tests and combined sensitivity, specificity, positive and negative likelihood ratio, diagnostic odds ratio, and area under the summary receiver operating characteristic curve will be performed by using Meta Disc1.4 software. Stata14.0 software will be used for sensitivity analysis and publication bias test.Results:The results of this systematic review will demonstrate the accuracy of the two elastography methods in the diagnosis of breast tumors.Discussion:The results will provide useful evidence for the comparison of the diagnostic accuracy of shear wave elastography and strain elastography in breast tumors.Other:This study was not funded. Register name: PROSPERO. Registration number: CRD42021251110.

Automatic defect detection for ultrasonic wave propagation imaging method using spatio-temporal convolution neural networks

Ultrasonic wave propagation imaging enables the detection of anomalies in various structures; hence, it has been applied as one of the promising techniques for damage identification in structural health monitoring (SHM). The interpretation of imaging data is vital to SHM; however, it relies significantly on expert subjective judgment, rendering the results vulnerable to human errors. Recent advances in the field of computer vision arising from the adoption of deep neural networks have resulted in new perspectives for substituting human roles in laborious data interpretation tasks. This paper presents an effective learning architecture that can characterize the ultrasonic wave propagation videos for automatic non-destructive inspection. The main contribution is threefold: 1. To the best of our knowledge, this is the first study to leverage video content analysis techniques to exploit ultrasonic wave propagation image series. Previous approaches that focused on the still wavefield images are likely to lose critical temporal information, thereby resulting in an inferior performance. 2. We devise a model that progressively aggregates both temporal and spatial information encoded in multiple adjacent snapshots of ultrasonic wave propagation motions for efficient data analysis. We presented the details regarding the system implementation and critical parameter settings. 3. The proposed approach is validated through extensive experimental comparisons with other state-of-the-art computer vision techniques on a real dataset which is publicly available. We hope that this study will encourage further investigations into video-based non-destructive data interpretation, not limited to ultrasonic signals.

Advanced Imaging Techniques for Newly Diagnosed and Recurrent Gliomas.

Management of gliomas following initial diagnosis requires thoughtful presurgical planning followed by regular imaging to monitor treatment response and survey for new tumor growth. Traditional MR imaging modalities such as T1 post-contrast and T2-weighted sequences have long been a staple of tumor diagnosis, surgical planning, and post-treatment surveillance. While these sequences remain integral in the management of gliomas, advances in imaging techniques have allowed for a more detailed characterization of tumor characteristics. Advanced MR sequences such as perfusion, diffusion, and susceptibility weighted imaging, as well as PET scans have emerged as valuable tools to inform clinical decision making and provide a non-invasive way to help distinguish between tumor recurrence and pseudoprogression. Furthermore, these advances in imaging have extended to the operating room and assist in making surgical resections safer. Nevertheless, surgery, chemotherapy, and radiation treatment continue to make the interpretation of MR changes difficult for glioma patients. As analytics and machine learning techniques improve, radiomics offers the potential to be more quantitative and personalized in the interpretation of imaging data for gliomas. In this review, we describe the role of these newer imaging modalities during the different stages of management for patients with gliomas, focusing on the pre-operative, post-operative, and surveillance periods. Finally, we discuss radiomics as a means of promoting personalized patient care in the future.

Use of Natural Language Processing (NLP) in Evaluation of Radiology Reports: An Update on Applications and Technology Advances

Natural language processing (NLP) is focused on the computer interpretation of human language and can be used to evaluate radiology reports and has demonstrated useful applications in essentially all aspects of medical imaging delivery: interpretation of imaging data, improving image acquisition, image analysis, and increasing efficiency of imaging services. This manuscript reviews general technologic approaches to NLP at a level hopefully understandable by clinical radiologists, discusses recent advancements in NLP techniques, and discusses current and potential applications of NLP in radiology.

Mapping cortical disease-burden at individual-level in frontotemporal dementia: implications for clinical care and pharmacological trials

Imaging studies of FTD typically present group-level statistics between large cohorts of genetically, molecularly or clinically stratified patients. Group-level statistics are indispensable to appraise unifying radiological traits and describe genotype-associated signatures in academic studies. However, in a clinical setting, the primary objective is the meaningful interpretation of imaging data from individual patients to assist diagnostic classification, inform prognosis, and enable the assessment of progressive changes compared to baseline scans. In an attempt to address the pragmatic demands of clinical imaging, a prospective computational neuroimaging study was undertaken in a cohort of patients across the spectrum of FTD phenotypes. Cortical changes were evaluated in a dual pipeline, using standard cortical thickness analyses and an individualised, z-score based approach to characterise subject-level disease burden. Phenotype-specific patterns of cortical atrophy were readily detected with both methodological approaches. Consistent with their clinical profiles, patients with bvFTD exhibited orbitofrontal, cingulate and dorsolateral prefrontal atrophy. Patients with ALS-FTD displayed precentral gyrus involvement, nfvPPA patients showed widespread cortical degeneration including insular and opercular regions and patients with svPPA exhibited relatively focal anterior temporal lobe atrophy. Cortical atrophy patterns were reliably detected in single individuals, and these maps were consistent with the clinical categorisation. Our preliminary data indicate that standard T1-weighted structural data from single patients may be utilised to generate maps of cortical atrophy. While the computational interpretation of single scans is challenging, it offers unrivalled insights compared to visual inspection. The quantitative evaluation of individual MRI data may aid diagnostic classification, clinical decision making, and assessing longitudinal changes.

ANALISIS BANJIR BANDANG DENGAN MENGGUNAKAN DATA CITRA SATELIT DI KABUPATEN LUWU UTARA (STUDI KASUS 13 JULI 2020)

Flash floods that hit parts of South Sulawesi on July 13, 2020, left 36 people dead, 18 people lost, 40 people missing, 58 injured, and a total of 14,483 people were affected. Around 14,000 people were evacuated to safer locations due to not having a home to live in. The disaster was caused by heavy rains. One of the areas affected was North Luwu. This phenomenon needs to be done research so that people can overcome the possibility of disasters that might occur. Air pressure and air temperature for 2 days before the occurrence of flash floods have the same relative patterns and anomalies occurred on the day of the incident. This research uses satellite image data interpretation method by processing the Himawari-8 infrared channel data and compared with rainfall data from the Andi Jemma meteorological station in North Luwu Regency on July 13, 2020. Interpretation and analysis of satellite imagery data shows the growth of strong convective clouds has the potential to cause rain in North Luwu around 07.00 UTC. So it was concluded that during a flash flood, unstable atmospheric conditions and the presence of Cumulonimbus (CB) cloud cover in the North Luwu region.

Detection method of tunnel lining voids based on guided anchoring mechanism

In tunnel construction engineering, the form of tunnel void diseases are complex and easily affected by the geographical environment. The traditional manual interpretation of image data has the characteristics of heavy workload, high probability of missing, and misjudgment. This paper constructs a convolution neural network that integrates the mechanism of guiding anchoring to detect tunnel voids. The network is composed of four parts: Feature extraction network extracts disease features from the enriched samples; Region proposal by guided anchoring join the generalized intersection over union (GIoU) evaluation criteria, and predict the shape of the anchor point through learning; The obtained feature maps are fixed in the region of interest pooling; Finally, the disease features are classified and bounding box regression. Compared with the existing target detection algorithm, the experimental results show that the improved network achieves an average classification accuracy of 92.74%, and the trained model has good generalization ability and robustness.

Evaluation and categorisation of individual patients based on white matter profiles: Single-patient diffusion data interpretation in neurodegeneration

The majority of radiology studies in neurodegenerative conditions infer group-level imaging traits from group comparisons. While this strategy is helpful to define phenotype-specific imaging signatures for academic use, the meaningful interpretation of single scans of individual subjects is more important in everyday clinical practice. Accordingly, we present a computational method to evaluate individual subject diffusion tensor data to highlight white matter integrity alterations. Fifty white matter tracts were quantitatively evaluated in 132 patients with amyotrophic lateral sclerosis (ALS) with respect to normative values from 100 healthy subjects. Fractional anisotropy and radial diffusivity alterations were assessed individually in each patient. The approach was validated against standard tract-based spatial statistics and further scrutinised by the assessment of 78 additional data sets with a blinded diagnosis. Our z-score-based approach readily detected white matter degeneration in individual ALS patients and helped to categorise single subjects with a 'blinded diagnosis' as likely ‘ALS’ or ‘control’. The group-level inferences from the z-score-based approach were analogous to the standard TBSS output maps. The benefit of the z-score-based strategy is that it enables the interpretation of single DTI datasets as well as the comparison of study groups. Outputs can be summarised either visually by highlighting the affected tracts, or, listing the affected tracts in a text file with reference to normative data, making it particularly useful for clinical applications. While individual diffusion data cannot be visually appraised, our approach provides a viable framework for single-subject imaging data interpretation.

Automated Pattern Recognition in Whole-Cardiac Cycle Echocardiographic Data: Capturing Functional Phenotypes with Machine Learning

•Pattern recognition is essential in cardiac imaging, but limited by operator variability. •Unsupervised machine learning (ML) can integrate whole-cardiac cycle echo data. •ML can recognize velocity and deformation patterns in single echo descriptors. •ML can integrate multiple echo descriptors to define patient phenogroups. •Automated pattern recognition has potential to aid everyday echo data interpretation. Background Echocardiography provides complex data on cardiac function that can be integrated into patterns of dysfunction related to the severity of cardiac disease. The aim of this study was to demonstrate the feasibility of applying machine learning (ML) to automate the integration of echocardiographic data from the whole cardiac cycle and to automatically recognize patterns in velocity profiles and deformation curves, allowing the identification of functional phenotypes. Methods Echocardiography was performed in 189 clinically managed patients with hypertension and 97 healthy individuals without hypertension. Speckle-tracking analysis of the left ventricle and atrium was performed, and deformation curves were extracted. Aortic and mitral blood pool pulsed-wave Doppler and mitral annular tissue pulsed-wave Doppler velocity profiles were obtained. These whole–cardiac cycle deformation and velocity curves were used as ML input. Unsupervised ML was used to create a representation of patients with hypertension in a virtual space in which patients are positioned on the basis of the similarity of their integrated whole–cardiac cycle echocardiography data. Regression methods were used to explore patterns of echocardiographic traces within this virtual ML-derived space, while clustering was used to define phenogroups. Results The algorithm captured different patterns in tissue and blood-pool velocity and deformation profiles and integrated the findings, yielding phenotypes related to normal cardiac function and others to advanced remodeling associated with pressure overload in hypertension. The addition of individuals without hypertension into the ML-derived space confirmed the interpretation of normal and remodeled phenotypes. Conclusions ML-based pattern recognition is feasible from echocardiographic data obtained during the whole cardiac cycle. Automated algorithms can consistently capture patterns in velocity and deformation data and, on the basis of these patterns, group patients into interpretable, clinically comprehensive phenogroups that describe structural and functional remodeling. Automated pattern recognition may potentially aid interpretation of imaging data and diagnostic accuracy. Echocardiography provides complex data on cardiac function that can be integrated into patterns of dysfunction related to the severity of cardiac disease. The aim of this study was to demonstrate the feasibility of applying machine learning (ML) to automate the integration of echocardiographic data from the whole cardiac cycle and to automatically recognize patterns in velocity profiles and deformation curves, allowing the identification of functional phenotypes. Echocardiography was performed in 189 clinically managed patients with hypertension and 97 healthy individuals without hypertension. Speckle-tracking analysis of the left ventricle and atrium was performed, and deformation curves were extracted. Aortic and mitral blood pool pulsed-wave Doppler and mitral annular tissue pulsed-wave Doppler velocity profiles were obtained. These whole–cardiac cycle deformation and velocity curves were used as ML input. Unsupervised ML was used to create a representation of patients with hypertension in a virtual space in which patients are positioned on the basis of the similarity of their integrated whole–cardiac cycle echocardiography data. Regression methods were used to explore patterns of echocardiographic traces within this virtual ML-derived space, while clustering was used to define phenogroups. The algorithm captured different patterns in tissue and blood-pool velocity and deformation profiles and integrated the findings, yielding phenotypes related to normal cardiac function and others to advanced remodeling associated with pressure overload in hypertension. The addition of individuals without hypertension into the ML-derived space confirmed the interpretation of normal and remodeled phenotypes. ML-based pattern recognition is feasible from echocardiographic data obtained during the whole cardiac cycle. Automated algorithms can consistently capture patterns in velocity and deformation data and, on the basis of these patterns, group patients into interpretable, clinically comprehensive phenogroups that describe structural and functional remodeling. Automated pattern recognition may potentially aid interpretation of imaging data and diagnostic accuracy.