Image Interpretation System Research Articles

Operational challenges and collaborative solutions in radiology image interpretation: perspectives from imaging departments in a low-resource setting.

Medical imaging's critical role in diagnosis requires prompt and precise image interpretation. Numerous radiology departments, especially in low-resourced settings, encounter challenges such as a shortage of radiologists in their operational setup. This study explored the perceptions of radiographers and radiologists from low-resourced departments in a single country regarding operational challenges and potential solutions in image interpretation. A qualitative approach was utilised, involving heads of departments, senior radiographers, and radiologists with a minimum of 5 years of experience, from three major state referral hospitals. Face-to-face, semi-structured interviews were conducted in November 2022, using an interview guide that included questions on the challenges encountered during image interpretation and the proposed solutions. Data analysis was conducted using Atlas.ti version 9.0, following the four-step content analysis method. All participants willingly provided consent to participate in the study. Ten participants, comprising two radiologists and eight radiographers participated in the study. The research identified three main themes: image interpretation pathways, image interpretation operational challenges and proposed solutions for image interpretation. In addition, a total of 10 subthemes were generated from the three main themes. The study revealed critical challenges and the need to explore the formal inclusion of radiographers in image interpretation, as a way to improve efficiency. However, a comprehensive assessment of the image interpretation system, encompassing radiographers' knowledge and competence, is recommended for context-specific, empirical-based modifications to enhance service provision.

Thematic interpretation of aerial photographs of forest areas based on conceptual modeling

Earth remote sensing (ERS) is one of the main sources of information about the earth's surface. With the development of unmanned aerial vehicles (UAVs), it became possible to take aerial photographs with high spatial resolution, which can be used to more accurately identify objects. The purpose of the study is to increase the reliability of aerial imagery interpretation by developing a methodology for automated image processing based on conceptual modeling. An analysis of the methods for thematic interpretation of images showed that none of them provides sufficient quality of segmentation without additional adjustment to the subject area. The technique of conceptual modeling of subject tasks is applied, which ensures the adequacy of syntactic representations (including various images) and makes it possible to control the logic of solving problems and reduce the number of errors at the stage of its software implementation. The description of the information system for thematic interpretation of images, which implements the developed methodology, is given. An assessment was made of the reliability of thematic interpretation of images of forest areas. It is shown that the developed technique allowed to improve the identification of forest objects on aerial photographs by 18.6%. The development of this technique can be carried out to determine the characteristics of identifiable trees: age, species, height, and wood stock.

A Behaviorally Evidence-based Method for Computing Spatial Comparisons of Image Scenarios

<p>Large amounts of noise and a lack of contextual domain knowledge lead to slow and inefficient cross-domain image learning. This paper proposes an image scenario spatial data classification model based on evidence-based behavioral logic, intervenes in image annotation through evidence-based dynamic knowledge graphs, and uses spatial similarity measurement to evaluate the effectiveness and robustness of the method. The results show that: 1) Organizing the dynamic knowledge graphs of contextual domain knowledge by behavioral logic can significantly improve the association efficiency of each model. 2) The calculation method of image scenario space comparison based on behavior evidence can decrypt the implicit knowledge of images and significantly improve the effectiveness of image scenario space interpretation. The research results are helpful to guide the design and implementation of cross-domain image interpretation systems and improve the efficiency of information sharing.</p> <p> </p>

Analisis Kesesuaian Lahan untuk Tanaman Jagung Menggunakan Citra Landsat 8 OLI/TIRS dan Sistem Informasi Geografis di Kabupaten Lampung Selatan

Analysis of land suitability for corn plants is needed to increase production in the development of the agricultural sector. South Lampung Regency is the largest corn producer in Lampung Province with production reaching 598,032 tonnes in 2016. The aim of the study was to determine the potential for land development for corn plants using Landsat 8 OLI/TIRS image interpretation and Geographic Information Systems in South Lampung Regency by assessing the quality of the land and to determine the level of land suitability conditions for corn crops based on planting seasons 1, 2 and 3 in South Lampung Regency. The analysis is carried out by the matching method or by giving a class assessment to each of the required parameters. The results show that the 1st planting season has the greatest level of land suitability compared to the other growing seasons, 73% very suitable to suitable. Meanwhile, Planting Season 2 had the lowest suitability level of 60% due to less stable temperatures due to the transition from rainy to dry seasons. And Planting Season 3 has a moderate suitability level of 66%, because this season the temperature has stabilized for cultivating corn. Keywords: Corn, Geographic Information System, Class Assessment, Land Suitability, Landsat 8 OLI/TIRS Imagery, Matching.

An autonomous X-ray image acquisition and interpretation system for assisting percutaneous pelvic fracture fixation.

Percutaneous fracture fixation involves multiple X-ray acquisitions to determine adequate tool trajectories in bony anatomy. In order to reduce time spent adjusting the X-ray imager's gantry, avoid excess acquisitions, and anticipate inadequate trajectories before penetrating bone, we propose an autonomous system for intra-operative feedback that combines robotic X-ray imaging and machine learning for automated image acquisition and interpretation, respectively. Our approach reconstructs an appropriate trajectory in a two-image sequence, where the optimal second viewpoint is determined based on analysis of the first image. A deep neural network is responsible for detecting the tool and corridor, here a K-wire and the superior pubic ramus, respectively, in these radiographs. The reconstructed corridor and K-wire pose are compared to determine likelihood of cortical breach, and both are visualized for the clinician in a mixed reality environment that is spatially registered to the patient and delivered by an optical see-through head-mounted display. We assess the upper bounds on system performance through in silico evaluation across 11 CTs with fractures present, in which the corridor and K-wire are adequately reconstructed. In post hoc analysis of radiographs across 3 cadaveric specimens, our system determines the appropriate trajectory to within 2.8 ± 1.3mm and 2.7 ± 1.8[Formula: see text]. An expert user study with an anthropomorphic phantom demonstrates how our autonomous, integrated system requires fewer images and lower movement to guide and confirm adequate placement compared to current clinical practice. Code and data are available.

The Principle and State-of-art Applications for CT Detector

Contemporarily, significant progress has been made in the development and application of image fusion technology in the field of diagnostic imaging. Image fusion is a fully automated image alignment and interpretation system that not only allows images with different characteristics to be displayed, interpreted, compared, and analyzed on the same platform, but also provides a fast and seamless pathway to clinical diagnosis and treatment. The purpose of this study was to explore the imaging theory, clinical advantages and limitations of CT. At the beginning of the study, the definition and development history of CT will be explained. Subsequently, in addition to imaging theory and key parameters, one can also understand the individual advantages and clinical application of CT. Afterwards, the limitations of CT will be mentioned, and the new technology of PET/CT will be introduced. As the most advanced medical imaging technology, PET/CT has improved the way of single PET and CT scanning, making them more widely used, making it the best combination of medical imaging and diagnostic technology. Overall, these results shed light on guiding further exploration of CT.

Development and validation of an abnormality-derived deep-learning diagnostic system for major respiratory diseases

Respiratory diseases impose a tremendous global health burden on large patient populations. In this study, we aimed to develop DeepMRDTR, a deep learning-based medical image interpretation system for the diagnosis of major respiratory diseases based on the automated identification of a wide range of radiological abnormalities through computed tomography (CT) and chest X-ray (CXR) from real-world, large-scale datasets. DeepMRDTR comprises four networks (two CT-Nets and two CXR-Nets) that exploit contrastive learning to generate pre-training parameters that are fine-tuned on the retrospective dataset collected from a single institution. The performance of DeepMRDTR was evaluated for abnormality identification and disease diagnosis on data from two different institutions: one was an internal testing dataset from the same institution as the training data and the second was collected from an external institution to evaluate the model generalizability and robustness to an unrelated population dataset. In such a difficult multi-class diagnosis task, our system achieved the average area under the receiver operating characteristic curve (AUC) of 0.856 (95% confidence interval (CI):0.843–0.868) and 0.841 (95%CI:0.832–0.887) for abnormality identification, and 0.900 (95%CI:0.872–0.958) and 0.866 (95%CI:0.832–0.887) for major respiratory diseases’ diagnosis on CT and CXR datasets, respectively. Furthermore, to achieve a clinically actionable diagnosis, we deployed a preliminary version of DeepMRDTR into the clinical workflow, which was performed on par with senior experts in disease diagnosis, with an AUC of 0.890 and a Cohen’s k of 0.746–0.877 at a reasonable timescale; these findings demonstrate the potential to accelerate the medical workflow to facilitate early diagnosis as a triage tool for respiratory diseases which supports improved clinical diagnoses and decision-making.

Categorized criteria for X-ray image interpretation for population-based gastric cancer screening and reconstruction of the X-ray image interpretation system for gastric cancer screening.

Categorized criteria for X-ray image interpretation for population-based gastric cancer screening and reconstruction of the X-ray image interpretation system for gastric cancer screening.

Beyond visual semantics: Exploring the role of scene text in image understanding

Images with visual and scene text content are ubiquitous in everyday life. However, current image interpretation systems are mostly limited to using only the visual features, neglecting to leverage the scene text content. In this paper, we propose to jointly use scene text and visual channels for robust semantic interpretation of images. We do not only extract and encode visual and scene text cues, but also model their interplay to generate a contextual joint embedding with richer semantics. The contextual embedding thus generated is applied to retrieval and classification tasks on multimedia images, with scene text content, to demonstrate its effectiveness. In the retrieval framework, we augment our learned text-visual semantic representation with scene text cues, to mitigate vocabulary misses that may have occurred during the semantic embedding. To deal with irrelevant or erroneous recognition of scene text, we also apply query-based attention to our text channel. We show how the multi-channel approach, involving visual semantics and scene text, improves upon state of the art.

POS0256 AUTOMATED CAPILLARY DETECTION AND IMAGE ANALYSIS SOFTWARE IN CAPILLAROSCOPY: CAPILLARY.IO

Background: A nailfold capillaroscopy procedure is a non-invasive, low-cost, and well-established examination that can be used to diagnose several rheumatic autoimmune diseases and support the necessary follow-up of patients. There are two main elements to nailfold capillaroscopy: image acquisition and image interpretation. Both of these can present challenges: we need to ensure that the best possible images are captured, and we need to define the enlarged capillaries, capillary loss and pericapillary hemorrhages objectively. We introduce Capillary.io, an automatic image reading system able to recognize capillaries in images obtained with any microscope, generate automatic measurements of each capillary and take advantage of this information to report capillary morphology. Together it allows a comprehensive analysis that is capable of producing detailed reports for each patient. Objectives: The primary outcome was general sensitivity and specificity, using images assessed by expert capillaroscopists as the gold standard. Methods: 6500 images previously analyzed by capillaroscopists from GREC were compared with Capillary.io. Capillary morphology (enlarged capillaries, tortuosities, ramifications, megacapillaries, hemorrhages) of each of the capillaries contained in each of the images was analyzed manually by at least one expert capillaroscopist. Subsequently, the automatic image interpretation system was used to fully automatically analyze each of the capillaries contained in each image and the results obtained were compared. Results: Overall, a total of 78.347 capillaries were compared, of which 47.734 were normal capillaries, 21.991 enlarged capillaries, 2672 megacapillaries, 8512 tortuosities, 1322 ramifications and 5149 hemorrhages. Capillary.io was able to detect 38.101 normal capillaries, 19.126 enlarged capillaries, 2389 megacapillaries, 5698 tortuosities, 718 ramifications and 3706 hemorraghes. Capillary.io presented a sensitivity (S) of 79.82% and a specificity (E) of 82% in the recognition of normal capillaries. The automatized system was able to recognize enlarged capillaries with a sensitivity of 86.97% and a specificity of 81.38%. Megacapillaries were detected with 89.41% sensitivity and 78.75% specificity. Similarly, the system was able to detect tortuosities (S 66.94%; E 67.71%), ramifications (S 54.34%; E 58.61%) and hemorrhages (S 71.36; E 73.97%). Conclusion: Capillary.io is a simple, easy-learning web-based system to get interpretation of nailfold capillaroscopic images. It may be a very useful tool to standardize the interpretation of capillaroscopic pictures and could provide great research in that field.

A morphology-based radiological image segmentation approach for efficient screening of COVID-19

Computer-aided radiological image interpretation systems can be helpful to reshape the overall workflow of the COVID-19 diagnosis process. This article describes an unsupervised CT scan image segmentation approach. This approach begins by performing a morphological reconstruction operation that is useful to remove the effect of the external disturbances on the infected regions and to locate different regions of interest precisely. The optimal size of the structuring element is selected using the Edge Content-based contrast matrix approach. After performing the opening by using the morphological reconstruction operation, further noise is eliminated using the closing-based morphological reconstruction operation. The original pixel space is restored and the obtained image is divided into some non-overlapping smaller blocks and the mean intensity value for each block is computed that is used as the local threshold value for the binarization purpose. It is preferable to manually determine the range of the infected region. If a region is greater than the upper bound then that region will be considered as an exceptional region and processed separately. Three standard metrics MSE, PSNR, and SSIM are used to quantify the outcomes. Both quantitative and qualitative comparisons prove the efficiency and real-life adaptability of this approach. The proposed approach is evaluated with the help of 400 different images and on average, the proposed approach achieves MSE 307.1888625, PSNR 23.7246505, and SSIM 0.831718459. Moreover, the comparative study shows that the proposed approach outperforms some of the standard methods and obtained results are encouraging to support the battle against the COVID-19.

Assessment of Rainfall-Induced Landslide Distribution Based on Land Disturbance in Southern Taiwan

This study explores the impact of rainfall on the followed-up landslides after a severe typhoon and the relationship between various rainfall events and the occurrence, scale, and regional characteristics of the landslides, including second landslides. Moreover, the influence of land disturbance was evaluated. The genetic adaptive neural network was used in combination with the texture analysis of the geographic information system for satellite image classification and interpretation to analyze land-use change and retrieve disaster records and surface information after five rainfall events from Typhoon Morakot (2009) to Typhoon Nanmadol (2011). The results revealed that except for extreme Morakot rains, the greater the degree of slope disturbance after rain, the larger the exposed slope. Extreme rainfall similar to Morakot strikes may have a greater impact on the bare land area than on slope disturbance. Moreover, the relationship between the bare land area and the index of land disturbance condition (ILDC) is positive, and the ratio of the bare land area to the quantity of bare land after each rainfall increases with the ILDC. With higher effective accumulative rainfall on the slope in the study area or greater slope disturbance, the landslide area at the second landslide point tended to increase.

Artificial Intelligence and Robotics in Regional Anaesthesia: Do they have a role?

Artificial Intelligence and Robotics in Regional Anaesthesia: Do they have a role?

Deep Learning for US Image Quality Assessment Based on Femoral Cartilage Boundary Detection in Autonomous Knee Arthroscopy.

Knee arthroscopy is a complex minimally invasive surgery that can cause unintended injuries to femoral cartilage or postoperative complications, or both. Autonomous robotic systems using real-time volumetric ultrasound (US) imaging guidance hold potential for reducing significantly these issues and for improving patient outcomes. To enable the robotic system to navigate autonomously in the knee joint, the imaging system should provide the robot with a real-time comprehensive map of the surgical site. To this end, the first step is automatic image quality assessment, to ensure that the boundaries of the relevant knee structures are defined well enough to be detected, outlined, and then tracked. In this article, a recently developed one-class classifier deep learning algorithm was used to discriminate among the US images acquired in a simulated surgical scenario on which the femoral cartilage either could or could not be outlined. A total of 38 656 2-D US images were extracted from 151 3-D US volumes, collected from six volunteers, and were labeled as "1" or as "0" when an expert was or was not able to outline the cartilage on the image, respectively. The algorithm was evaluated using the expert labels as ground truth with a fivefold cross validation, where each fold was trained and tested on average with 15 640 and 6246 labeled images, respectively. The algorithm reached a mean accuracy of 78.4% ± 5.0, mean specificity of 72.5% ± 9.4, mean sensitivity of 82.8% ± 5.8, and mean area under the curve of 85% ± 4.4. In addition, interobserver and intraobserver tests involving two experts were performed on an image subset of 1536 2-D US images. Percent agreement values of 0.89 and 0.93 were achieved between two experts (i.e., interobserver) and by each expert (i.e., intraobserver), respectively. These results show the feasibility of the first essential step in the development of automatic US image acquisition and interpretation systems for autonomous robotic knee arthroscopy.

Analisis Keselarasan Pemanfaatan Ruang dengan Rencana Pola Ruang dan Pengendaliannya di Kota Jakarta Timur

The rapid development in East Jakarta City has led to increased land requirements and causes unalignment of land use. Research objective are to analyze the alignment of spatial usage with the spatial plan and its control in East Jakarta City. The research methods used are spatial analysis based on image interpretation and Geographic Information System, descriptive statistic for seeking influence factors, description analysis for spatial control usage and its guidance. Results analysis for 6 (six) existing land use showed that alignment of existing land use has area of 12.430,5 ha or 76.5 % and unalignment of existing land use has area 3.812,7 ha or 23.5%. Factors that influence unalignment of existing land use, namely economic needs, housing needs, distance from the economic center, distance from the road, lack of socialization to the community, population density and the presence of public facilities. Analysis of the implementation of spatial usage control shows that the East Jakarta City Government has implemented 4 spatial usage control instruments namely the Spatial Detail Plan along with its Zoning Regulations, licensing, providing incentives and disincentives and sanctions but the implementation is still considered not maximally. Directives for improving spatial usage are carried out by maximizing and improving the implementation of four instruments for controlling spatial usage.

Automated measurement of bone scan index from a whole-body bone scintigram

PurposeWe propose a deep learning-based image interpretation system for skeleton segmentation and extraction of hot spots of bone metastatic lesion from a whole-body bone scintigram followed by automated measurement of a bone scan index (BSI), which will be clinically useful.MethodsThe proposed system employs butterfly-type networks (BtrflyNets) for skeleton segmentation and extraction of hot spots of bone metastatic lesions, in which a pair of anterior and posterior images are processed simultaneously. BSI is then measured using the segmented bones and extracted hot spots. To further improve the networks, deep supervision (DSV) and residual learning technologies were introduced.ResultsWe evaluated the performance of the proposed system using 246 bone scintigrams of prostate cancer in terms of accuracy of skeleton segmentation, hot spot extraction, and BSI measurement, as well as computational cost. In a threefold cross-validation experiment, the best performance was achieved by BtrflyNet with DSV for skeleton segmentation and BtrflyNet with residual blocks. The cross-correlation between the measured and true BSI was 0.9337, and the computational time for a case was 112.0 s.ConclusionWe proposed a deep learning-based BSI measurement system for a whole-body bone scintigram and proved its effectiveness by threefold cross-validation study using 246 whole-body bone scintigrams. The automatically measured BSI and computational time for a case are deemed clinically acceptable and reliable.

Editors Note

In this edition, covering a variety of topics across medical and dental research, as well as materials science and historical epidemiology, we see a strong representation of the value of international and interspecialty collaboration. This issue includes many such collaborative projects, including several projects that utilise developments in technology to further desired medical outcomes, combining the skills of experts in fields as varied as image and gesture interpretation research, systems and information engineering, rehabilitation engineering, medicine and occupational therapy.

P152: Point of care biliary ultrasound in the emergency department (BUSED): implications for surgical referral and emergency department wait times

Introduction: Patients with uncomplicated biliary disease frequently present to the emergency department for assessment. To improve bedside clinical decision making many emergency medicine physicians have pursued specialized training to perform point of care biliary ultrasound in the emergency department (BUSED). The purpose of this study was to determine the usefulness of BUSED in predicting the need for surgical consultation and intervention for biliary disease. Methods: A retrospective study of 283 consecutive patients visiting the emergency department who received a BUSED scan from December 1, 2016 to July 16, 2017. Physician interpretations of the BUSED scans were collected from the electronic image storage and interpretation system. Additional data was collected from the electronic health record including lab values, the subsequent use of diagnostic imaging, and outcomes data including disposition, surgical consultation or intervention, and 28 day follow up for representation or complication. Descriptive statistics and logistic regression were performed. Results: Of the patients who received a biliary POCUS scan, 29% were referred to general surgery, and 43% of those referred proceeded to eventual cholecystectomy. Factors found to be independently predictive of surgical intervention on point of care BUSED scans included presence of gall stones (OR 13.01, 95%CI 5.02 to 27.1) and increased gallbladder wall thickness (OR 6.01, 95%CI 1.7 to 11.1). A total of 30% of patients receiving BUSED required at least one additional, radiology based imaging test (CT or diagnostic US). Average emergency department length of stay was substantially longer for those who required additional imaging as compared to those who were able to be diagnosed by BUSED alone (16.1 versus 5.2 hours, 10.9 hours 95%CI 10.6 11.2, p<0.05) . Conclusion: Point of care biliary ultrasound performed by emergency physicians provides timely access to diagnostic information. Positive findings of gall stones and increased gall bladder wall thickness are highly predictive of the need for surgical intervention. Future, prospective studies are warranted to determine if point of care sonography is sufficient to proceed to surgery in select cases of uncomplicated biliary disease.

Predicting cognitive decline with deep learning of brain metabolism and amyloid imaging

For effective treatment of Alzheimer’s disease (AD), it is important to identify subjects who are most likely to exhibit rapid cognitive decline. We aimed to develop an automatic image interpretation system based on a deep convolutional neural network (CNN) which can accurately predict future cognitive decline in mild cognitive impairment (MCI) patients using flurodeoxyglucose and florbetapir positron emission tomography (PET). PET images of 139 patients with AD, 171 patients with MCI and 182 normal subjects obtained from Alzheimer’s Disease Neuroimaging Initiative database were used. Deep CNN was trained using 3-dimensional PET volumes of AD and normal controls as inputs. Manually defined image feature extraction such as quantification using predefined region-of-interests was unnecessary for our approach. Furthermore, it used minimally processed images without spatial normalization which has been commonly used in conventional quantitative analyses. Cognitive outcome of MCI subjects was predicted using this network. The prediction accuracy of the conversion of mild cognitive impairment to AD was compared with the conventional feature-based quantification approach. Accuracy of prediction (84.2%) for conversion to AD in MCI patients outperformed conventional feature-based quantification approaches. ROC analyses revealed that performance of CNN-based approach was significantly higher than that of the conventional quantification methods (p < 0.05). Output scores of the network were strongly correlated with the longitudinal change in cognitive measurements (p < 0.05). These results show the feasibility of deep learning as a practical tool for developing predictive neuroimaging biomarker.

Fast target detection in radar images using Rayleigh Mixtures and Summed Area Tables

As the first step of automatic image interpretation systems, automatic detection of targets should be accurate and fast. For Synthetic Aperture Radar (SAR) images, Constant False Alarm Rate (CFAR) is the most popular framework used for target detection. In CFAR, modeling of the clutter is crucial since the decision threshold is calculated based on this model. In this study, we propose to model the background statistics using a Rayleigh Mixture (RM) model. Such an approach facilitates modeling of complex statistics, including but not limited to those involved in heavy tailed distributions, which are shown to be good fits especially for high resolution SAR images. We also propose an efficient method to evaluate CFAR thresholds according to the proposed model by use of Summed Area Tables (SAT). SAT provides a remarkable efficiency as the Rayleigh distribution is represented by only one parameter that can be estimated using simple moments. Tiling and parallel implementation is also utilized for fast computation of results. The outcome is a highly-accurate, extremely fast, and adaptive target detection approach that can be seamlessly used with a variety of complex SAR scenes. Our experiments compare the proposed approach with existing target detection methods and demonstrate its effectiveness as well as the benefits it provides.