透视图像亮度自动控制

X射线机所配置的影像增强器电视系统在进行人体透视时,当人体部位密度改变,监视器荧屏图像亮度将随之发生变化。为了保持监视器荧屏图像亮度的稳定性,采用透视图像亮度自动控制是十分必要的,其有助于保证X射线成像质量,提高临床诊断效率并给临床操作带来很大的便利。     

Segmentation of organs-at-risk in cervical cancer CT images with a convolutional neural network

PurposeWe introduced and evaluated an end-to-end organs-at-risk (OARs) segmentation model that can provide accurate and consistent OARs segmentation results in much less time.MethodsWe collected 105 patients’ Computed Tomography (CT) scans that diagnosed locally advanced cervical cancer and treated with radiotherapy in one hospital. Seven organs, including the bladder, bone marrow, left femoral head, right femoral head, rectum, small intestine and spinal cord were defined as OARs. The annotated contours of the OARs previously delineated manually by the patient’s radiotherapy oncologist and confirmed by the professional committee consisted of eight experienced oncologists before the radiotherapy were used as the ground truth masks. A multi-class segmentation model based on U-Net was designed to fulfil the OARs segmentation task. The Dice Similarity Coefficient (DSC) and 95th Hausdorff Distance (HD) are used as quantitative evaluation metrics to evaluate the proposed method.ResultsThe mean DSC values of the proposed method are 0.924, 0.854, 0.906, 0.900, 0.791, 0.833 and 0.827 for the bladder, bone marrow, femoral head left, femoral head right, rectum, small intestine, and spinal cord, respectively. The mean HD values are 5.098, 1.993, 1.390, 1.435, 5.949, 5.281 and 3.269 for the above OARs respectively.ConclusionsOur proposed method can help reduce the inter-observer and intra-observer variability of manual OARs delineation and lessen oncologists’ efforts. The experimental results demonstrate that our model outperforms the benchmark U-Net model and the oncologists’ evaluations show that the segmentation results are highly acceptable to be used in radiation therapy planning.     

MtreeRing: An R package with graphical user interface for automatic measurement of tree ring widths using image processing techniques

Accurate measurements of ring-width series are essential for dendrochronological analyses. We present an R package MtreeRing for ring-width measurements on scanned digital images. A morphological alternate sequential filter is used for noise reduction in the original image. Ring boundaries are determined by the steepest negative slopes in the light reflectance of latewood-earlywood transitions. To automatically identify tree rings, the package provides three alternative methods (watershed-based segmentation, Canny edge detector, and a linear detection algorithm), each with advantages and disadvantages and suited to different wood anatomical features. The user can also manually mark tree rings on species with complex anatomical structures. The arcs of inner-rings and angles of successive inclined ring boundaries are used to correct ring-width series. Differences in ring-width measurements between MtreeRing and WinDENDRO in a given coniferous species (Larix gmelinii) were assessed, and no significant difference between programs was found. Furthermore, the package provides an R-based web application which was developed using the Shiny framework. This beginner-friendly application allows viewing and interacting with tree ring images. It requires no programming experience and can run on either a local computer or a remote server.     

A review of automatic recognition technology for bird vocalizations in the deep learning era

Birds are considered critical indicators of ecosystem condition. Automatic recording devices have emerged as a trending tool to assist field observations, contributing to biodiversity monitoring on large spatio-temporal scales. However, manually processing huge volumes of recordings is challenging. Consequently, there has been a growing interest in automatic bird vocalization recognition in recent years. Automatic bird vocalization recognition technology has advanced from classical pattern recognition to deep learning (DL), with significantly improved recognition performance. This paper reviews related works on DL-based automatic bird vocalization recognition technology in the last decade. In this review, we present the current state of research in the three key areas of pre-processing, feature extraction and recognition methods involved in automatic bird vocalization recognition. The related datasets, evaluation metrics and software are also summarized. Finally, existing challenges along with opportunities for future work are highlighted. We conclude that, while DL-based automatic bird vocalization recognition has made recent advances in specific species, more robust denoising approaches, larger public datasets, and stronger generalization capabilities of feature extraction and recognition are required to achieve reliable and general bird recognition in the wild. We expect that this review will serve as a firm foundation for new researchers working in the field of DL-based automatic bird vocalization recognition technologies, as well as become an insightful guide for computer science and ecology experts.     

Novel knowledge-based treatment planning model for hypofractionated radiotherapy of prostate cancer patients

PurposeTo demonstrate the strength of an innovative knowledge-based model-building method for radiotherapy planning using hypofractionated, multi-target prostate patients.Material and methodsAn initial RapidPlan model was trained using 48 patients who received 60 Gy to prostate (PTV60) and 44 Gy to pelvic nodes (PTV44) in 20 fractions. To improve the model's goodness-of-fit, an intermediate model was generated using the dose-volume histograms of best-spared organs-at-risk (OARs) of the initial model. Using the intermediate model and manual tweaking, all 48 cases were re-planned. The final model, trained using these re-plans, was validated on 50 additional patients. The validated final model was used to determine any planning advantage of using three arcs instead of two on 16 VMAT cases and tested on 25 additional cases to determine efficacy for single-PTV (PTV60-only) treatment planning.ResultsFor model validation, PTV V_95% of 99.9% was obtained by both clinical and knowledge-based planning. D_1% was lower for model plans: by 1.23 Gy (PTV60, CI = [1.00, 1.45]), and by 2.44 Gy (PTV44, CI = [1.72, 3.16]). OAR sparing was superior for knowledge-based planning: ΔD_mean = 3.70 Gy (bladder, CI = [2.83, 4.57]), and 3.22 Gy (rectum, CI = [2.48, 3.95]); ΔD_2% = 1.17 Gy (bowel bag, CI = [0.64, 1.69]), and 4.78 Gy (femoral heads, CI = [3.90, 5.66]). Using three arcs instead of two, improvements in OAR sparing and PTV coverage were statistically significant, but of magnitudes < 1 Gy. The model failed at reliable DVH predictions for single PTV plans.ConclusionsOur knowledge-based model delivers efficient, consistent plans with excellent PTV coverage and improved OAR sparing compared to clinical plans.     

Knowledge-based automatic optimization of adaptive early-regression-guided VMAT for rectal cancer

PurposeTo implement a knowledge-based (KB) optimization strategy to our adaptive (ART) early-regression guided boosting technique in neo-adjuvant radio-chemotherapy for rectal cancer.Material and methodsThe protocol consists of a first phase delivering 27.6 Gy to tumor/lymph-nodes (2.3 Gy/fr-PTV₁), followed by the ART phase concomitantly delivering 18.6 Gy (3.1 Gy/fr) and 13.8 Gy (2.3 Gy/fr) to the residual tumor (PTV_ART) and to PTV₁ respectively. PTV_ART is obtained by expanding the residual GTV, as visible on MRI at fraction 9. Forty plans were used to generate a KB-model for the first phase using the RapidPlan tool. Instead of building a new model, a robust strategy scaling the KB-model to the ART phase was applied. Both internal and external validation were performed for both phases: all automatic plans (RP) were compared in terms of OARs/PTVs parameters against the original plans (RA).ResultsThe resulting automatic plans were generally better than or equivalent to clinical plans. Of note, V_30Gy and V_40Gy were significantly improved in RP plans for bladder and bowel; gEUD analysis showed improvement for KB-modality for all OARs, up to 3 Gy for the bowel.ConclusionsThe KB-model generated for the first phase was robust and it was also efficiently adapted to the ART phase. The performance of automatically generated plans were slightly better than the corresponding manual plans for both phases.     

Morphological quantification of filamentous fungal development using membrane immobilization and automatic image analysis

Mycelial morphology is a critically important process property in industrial fermentations of filamentous micro-organisms, as particular phenotypes are associated with maximum productivity. However, the accurate quantification of complex morphologies still represents a significant challenge in elucidating this relationship. A system has been developed for high-resolution characterisation of filamentous fungal growth on a solid substrate, using membrane immobilization and fully-automatic plug-ins developed for the public domain, Java-based, image-processing software, ImageJ. The system has been used to quantify the microscopic development of Aspergillus oryzae on malt agar, by measuring spore projected area and circularity, the total length of a hyphal element, the number of tips per element, and the hyphal growth unit. Two different stages of growth are described, from the swelling of a population of conidiospores up to fully developed, branched hyphae 24 h after inoculation. Spore swelling expressed as an increase in mean equivalent spore diameter was found to be approximately linear with time. Widespread germination of spores was observed by 8 h after inoculation. From approximately 12 h, the number of tips was found to increase exponentially. The specific growth rate of a population of hyphae was calculated as approximately 0.24–0.27 h⁻¹. A wide variation in growth kinetics was found within the population. The robustness of the image-analysis system was verified by testing the effect of small variations in the input data.     

Ovitrap Monitor - Online application for counting mosquito eggs and visualisation toolbox in support of health services

Ovitraps are a widely used method for mosquito detection and monitoring, especially Aedes mosquitoes. Eggs present in ovitraps must be routinely counted to generate up-to-date information on potential spread of mosquito-borne diseases. This task is tedious, time consuming and prone to errors if done manually by eye counting. In this contribution, we introduce the Ovitrap Monitor, an online open source and user-friendly integrated application that semi-automatically counts mosquito eggs from low-medium resolution mobile phone pictures. A high correlation was found between counts performed manually by a technician and those obtained with the app using an extensive dataset of more than 750 ovitrap pictures. The application features an intuitive user interface and time-series plots and maps to facilitate data flow and speed up evidence-based decision-making within health organisations battling mosquito-borne diseases. Besides being open source, the Ovitrap Monitor is also backed by test data to guarantee its implementation through benchmarking and enforce research in the public health field.     

Development of an automatic ROI setting program for input function determination in 99mTc-ECD non-invasive cerebral blood flow quantification

Non-invasive quantitative measurements are useful for clinical study as these are simple and pain-free procedures. A new non-invasive semi-automatic quantitative measurement method, the improved brain uptake ratio (IBUR) method using ^99mTc-ECD SPECT, has recently been reported. If an automatic ROI setting algorithm could be developed to determine the input function for the IBUR method, analysis of regional cerebral blood flow (rCBF) can be completed within a few min without recourse to complex techniques, through a fully automatic rCBF analysis program. The purpose of this study was to develop an automatic input function determination program for ^99mTc-ECD non-invasive cerebral blood flow quantification and to confirm the feasibility of use of this program.The images of 15 consecutive patients who underwent both ^99mTc-ECD chest RI angiography and SPECT examinations were used for development of the automatic arterial input function program. The images of 69 consecutive patients were used for validation of the program.The coincidence ratio between the ROI automatic method and the manual setting method was 98%. The mean difference in the ROI location was ±6.4 mm in the X direction and ±8.6 mm in the Y direction. Individual rCBF values obtained using these independent techniques were also reasonably well correlated (r = 0.95). The total time for the IBUR analysis using the automatic method is 2–3 min as compared to 20–30 min for the current analysis method. This technique improves the throughput of nuclear medical examinations.