Approach For Delineation Research Articles

Evaluating Regional Variability in Road Closure Outcomes Due to Rainfall: a Logistic Regression Approach

Abstract. This study investigated the probability of road closure due to flooding. Logistic regression model was developed using the road closure data and the daily rainfall data for Houston, TX, USA during 2017 and 2018. The road network was further divided into flood prone zones. The spatial analysis revealed that the rainfall at the road segment level could be sufficiently represented by that recorded by nearest sensors. Within a 4 d window, the rainfall in the current day and 3 d prior played a more influential role in predicting road closure. The differential outcomes due to distinct regional features were explained. Finally, a watershed delineation approach substantially improved the model's predictive power and sensitivity.

Recent strategies in target identification of natural products: Exploring applications in chronic inflammation and beyond.

Natural products are a treasure trove for drug discovery, especially in the areas of infection, inflammation and cancer, due to their diverse bioactivities and complex, and varied structures. Chronic inflammation is closely related to many diseases, including complex diseases such as cancer and neurodegeneration. Improving target identification for natural products contributes to elucidating their mechanism of action and clinical progress. It also facilitates the discovery of novel druggable targets and the elimination of undesirable ones, thereby significantly enhancing the productivity of drug discovery and development. Moreover, the rise of polypharmacological strategies, considered promising for the treatment of complex diseases, will further increase the demand for target deconvolution. This review underscores strategies for identifying natural product targets (NPs) in the context of chronic inflammation over the past 5 years. These strategies encompass computational methodologies for early target discovery and the anticipation of compound binding sites, proteomics-driven approaches for target delineation and experimental biology techniques for target validation and comprehensive mechanistic exploration.

Exploration of the Urbanization Process and Its Impact on Vegetation in 125 Resource-Based Cities in China and Comparison with Other Cities

Resource-based cities (RBCs) in China are at a historic juncture in their transformative development. Observing and assessing the role of the resource curse in urban expansion and greening is crucial for the sustainable development of these cities. This study proposes a new framework to extract urban boundary data from 2000 to 2020 in China. Utilizing these data, we analyzed differences in urban expansion intensity and urban vegetation cover between 125 RBCs and 223 non-RBCs. We found the following: (1) While urban areas in China experienced steady growth from 2000 to 2020, the urban area expansion rates of RBCs lagged behind those non-RBCs located in the same geographical areas except in South China, with the lowest annual expansion rate of 1.18% occurring in the Northeast. (2) Although the existing urban areas in some cities show a greening trend, both existing and new urban areas in China are predominantly characterized overall by browning. Surprisingly, RBCs exhibit a stronger greening trend than non-RBCs, particularly in Northwestern China. (3) There is a nuanced interplay and coexistence between resource dependency and urban expansion, with a specific negative correlation when resource dependency is very high or very low. This study provides a novel method and approach for urban boundary delineation. It offers new insights into the developmental characteristics of RBCs, enriching the theoretical framework of resource curse research and supporting the green development of RBCs with robust data.

A model integration approach for stratigraphic boundary delineation based on local data augmentation

Identification of stratigraphic boundaries is a fundamental task in the seismic interpretation of oil and gas reservoir locations. When employing deep learning techniques to interpret stratigraphic boundaries, insufficient training data and sample imbalances are common challenges affecting model training. In regions with intricate geological structural changes, conventional deep-learning segmentation algorithms, such as U-Net often struggle to accurately capture the features of complex local structures. To address these limitations, we propose a model integration approach that incorporates global and local uneven-type stratigraphic data augmentation to enhance the accuracy of stratigraphic boundary identification in uneven-type regions. To address the problems of class imbalance and insufficient complex variation samples, we adopted a strategy of separately training global and local data and integrating predictions, thereby handling the disparity between uneven-type and flat-type stratigraphic data during model training. By testing the Netherlands F3 dataset with sparsely labeled profiles, it was demonstrated that the proposed method can effectively improve the delineation accuracy of stratigraphic boundaries compared to the benchmark U-Net model.

A spatially-explicit sensitivity analysis of urban definitions: Uncovering implicit assumptions in the Degree of Urbanisation

We introduce a spatially-explicit sensitivity framework to uncover potential biases in urban delineation approaches. Our starting point is that there is no broadly shared agreement on how to define or delineate urban areas, neither in terms of methods nor in terms of thresholds or criteria. Deciding on delineation criteria thus inevitably involves making certain assumptions that may unwittingly reproduce urban realities experienced by those expressing them, and have spatially unequally distributed implications. Understanding how specific criterion choices shape our understanding of ‘the urban’ and how, why, and – especially – where a definition leads to specific sensitivities is therefore key, especially when the definition is utilised beyond its intended application. Our framework to uncover these sensitivities is spatially explicit in the sense that it does not rely on aggregate statistics but instead focuses on the sensitivity of the ‘urban’ classification of individual spatial units at the finest spatial granularity. Applying the framework to the definition of the Degree of Urbanisation reveals that sensitivity is indeed not equally distributed across the world. Certain regions (e.g., areas around Dallas – Fort Worth) and specific types of urbanisation (e.g., desakota regions in Pacific Asia) exhibit higher sensitivity than others. We discuss how these sensitivities may embody certain implicit assumptions in the definition, and examine their broader theoretical implications.

Introducing a tailored site delineation approach to optimize the design of managed aquifer recharge surface spreading infrastructure

Managed aquifer recharge (MAR) is an emerging solution to effectively replenish overused groundwater resources, but high associated costs often hinder its uptake. There are several parameters to consider when determining the cost of MAR, including recharge types, scale of MAR schemes, land acquisition, operation and maintenance period, and the hydrogeological setting. Hydraulic conductivity and its spatial variability are the most significant hydrogeological parameters for predicting infiltration rates at MAR sites, but limited data availability makes accurate predictions often challenging. Hence, it is essential to increase data availability to optimize MAR efficiency and to better assess its economic performance. Therefore, a novel approach for MAR planning is presented in this article to efficiently enhance the data availability during the conceptual stage of MAR planning by combining site exploration, spatially resolved infiltration rate estimation, and cost analysis. The approach was applied at an actual MAR site in Vincenza, Italy, and incorporated an electromagnetic induction survey and direct push profiling as across-scale investigation methods to enhance site delineation and subsequent parameterization of the identified zones using laboratory analysis of collected samples. This MAR site is particularly suitable for its heterogeneous subsurface structure and potential to counteract groundwater depletion. The performed investigation resulted in an image of the lateral variation of subsurface conditions and allowed the delineation of three zones with different infiltration behavior, with 99.5% of infiltration occurring in just two zones, representing only 22.1% of the site. In this particular scenario, the overall cost per cubic meter of recharged water can be reduced by 59.1% by identifying and eliminating unfavorable zones. This study provides scientists and practitioners with a useful tool for MAR planning that can be applied to a wide range of sites with complex geology, thereby reducing water costs, minimizing the environmental impact of infrastructure construction, and reducing land use conflicts.

PKSEA-Net: A prior knowledge supervised edge-aware multi-task network for retinal arteriolar morphometry

Retinal fundus images serve as a non-invasive modality to obtain information pertaining to retinal vessels through fundus photography, thereby offering insights into cardiovascular and cerebrovascular diseases. Retinal arteriolar morphometry has emerged as the most convenient and fundamental clinical methodology in the realm of patient screening and diagnosis. Nevertheless, the analysis of retinal arterioles is challenging attributable to imaging noise, stochastic fuzzy characteristics, and blurred boundaries proximal to blood vessels. In response to these limitations, we introduce an innovative methodology, named PKSEA-Net, which aims to improve segmentation accuracy by enhancing the perception of edge information in retinal fundus images. PKSEA-Net employs the universal architecture PVT-v2 as the encoder, complemented by a novel decoder architecture consisting of an Edge-Aware Block (EAB) and a Pyramid Feature Fusion Module (PFFM). The EAB block incorporates prior knowledge for supervision and multi-query for multi-task learning, with supervision information derived from an enhanced Full Width at Half Maximum (FWHM) algorithm and gradient map. Moreover, PFFM efficiently integrates multi-scale features through a novel attention fusion method. Additionally, we have collected a Retinal Cross-Sectional Vessel (RCSV) dataset derived from approximately 200 patients in Quzhou People’s Hospital to serve as the benchmark dataset. Comparative evaluations with several state-of-the-art (SOTA) networks confirm that PKSEA-Net achieves exceptional experimental performance, thereby establishing its status as a SOTA approach for precise boundary delineation and retinal vessel segmentation.

"Integrated Geospatial Solutions for Enhanced Kukadi River Basin Delineation using Remote Sensing and GIS Techniques"

This study offers an innovative approach to enhance the precision and accuracy of the delineation of the Kukadi River Basin by integrating cutting-edge geospatial technologies, particularly Remote Sensing (RS) and Geographic Information System (GIS) methods. Due to its varied topography and land cover, the Kukadi River Basin play a crucial hydrological entity in the area which presents difficulties for standard delineation techniques. High-resolution satellite imagery is used in the study to provide exact information on land cover in the Kukadi River Basin. The work aims to extract comprehensive land cover data by utilizing spectral analysis and image classification techniques, which are part of RS. In order to improve the delineation of the Basin borders, this data is seamlessly integrated with GIS spatial analytic tools, which also incorporate digital elevation models, slope, and other terrain metrics. The research findings demonstrate the possibility of combining RS and GIS technology to overcome the challenges associated with varying terrains, hence furthering the improvement of Basin delineation approaches. The results provide a useful and dependable paradigm for defining river Basin in a variety of geographical contexts, which is important for water resource management, Environmental Conservation, Land use planning, decision-makers, researchers, and practitioners working on hydrological studies.

Re-thinking rurality: Towards a new research approach and rural-urban spatial gradient establishment in Serbia

The research on the rural-urban gradient in Serbia has been reinvigorated following the debate on rural bias. Rural-urban classification is essential in observing spatial, demographic, socio-economic, and environmental processes. There is no consensus regarding the definition of rural and rurality and it should not be observed separately from urban issues. In Serbia, it is still a matter of dispute. Rural is most often observed through the urban–rural dichotomy and “left behind” for decades. This paper strives for a new theoretical understanding of the rurality in Serbia by introducing rural heterogeneity and ‘continuing’ rurality, considering the transformative phase they are affected by. It induces a process of rethinking what rurality in Serbia is and offers an innovative methodological approach for rural delineation and settlement classification based on three main pillars: settlement patterns, functional patterns, and local relevance. The analysis is conducted on the lowest administrative level in Central Serbia. Five types of settlements are identified: two urban, and three characterized by rural features. The combination of abstractions and the methodological opportunities introduced by rethinking rurality offers geographers and practitioners a powerful methodological tool for the understanding of rural transformations and transitional spatial forms, which underpins planning and policy-making in Serbia.

Generalising electrocardiogram detection and delineation: training convolutional neural networks with synthetic data augmentation.

Extracting beat-by-beat information from electrocardiograms (ECGs) is crucial for various downstream diagnostic tasks that rely on ECG-based measurements. However, these measurements can be expensive and time-consuming to produce, especially for long-term recordings. Traditional ECG detection and delineation methods, relying on classical signal processing algorithms such as those based on wavelet transforms, produce high-quality delineations but struggle to generalise to diverse ECG patterns. Machine learning (ML) techniques based on deep learning algorithms have emerged as promising alternatives, capable of achieving similar performance without handcrafted features or thresholds. However, supervised ML techniques require large annotated datasets for training, and existing datasets for ECG detection/delineation are limited in size and the range of pathological conditions they represent. This article addresses this challenge by introducing two key innovations. First, we develop a synthetic data generation scheme that probabilistically constructs unseen ECG traces from "pools" of fundamental segments extracted from existing databases. A set of rules guides the arrangement of these segments into coherent synthetic traces, while expert domain knowledge ensures the realism of the generated traces, increasing the input variability for training the model. Second, we propose two novel segmentation-based loss functions that encourage the accurate prediction of the number of independent ECG structures and promote tighter segmentation boundaries by focusing on a reduced number of samples. The proposed approach achieves remarkable performance, with a -score of 99.38% and delineation errors of ms and ms for ECG segment onsets and offsets across the P, QRS, and T waves. These results, aggregated from three diverse freely available databases (QT, LU, and Zhejiang), surpass current state-of-the-art detection and delineation approaches. Notably, the model demonstrated exceptional performance despite variations in lead configurations, sampling frequencies, and represented pathophysiology mechanisms, underscoring its robust generalisation capabilities. Real-world examples, featuring clinical data with various pathologies, illustrate the potential of our approach to streamline ECG analysis across different medical settings, fostered by releasing the codes as open source.

Characterization of Polarimetric Properties in Various Brain Tumor Types Using Wide-Field Imaging Mueller Polarimetry.

Neuro-oncological surgery is the primary brain cancer treatment, yet it faces challenges with gliomas due to their invasiveness and the need to preserve neurological function. Hence, radical resection is often unfeasible, highlighting the importance of precise tumor margin delineation to prevent neurological deficits and improve prognosis. Imaging Mueller polarimetry, an effective modality in various organ tissues, seems a promising approach for tumor delineation in neurosurgery. To further assess its use, we characterized the polarimetric properties by analysing 45 polarimetric measurements of 27 fresh brain tumor samples, including different tumor types with a strong focus on gliomas. Our study integrates a wide-field imaging Mueller polarimetric system and a novel neuropathology protocol, correlating polarimetric and histological data for accurate tissue identification. An image processing pipeline facilitated the alignment and overlay of polarimetric images and histological masks. Variations in depolarization values were observed for grey and white matter of brain tumor tissue, while differences in linear retardance were seen only within white matter of brain tumor tissue. Notably, we identified pronounced optical axis azimuth randomization within tumor regions. This study lays the foundation for machine learning-based brain tumor segmentation algorithms using polarimetric data, facilitating intraoperative diagnosis and decision making.

SEResUTer: a deep learning approach for accurate ECG signal delineation and atrial fibrillation detection

Objective. Accurate detection of electrocardiogram (ECG) waveforms is crucial for computer-aided diagnosis of cardiac abnormalities. This study introduces SEResUTer, an enhanced deep learning model designed for ECG delineation and atrial fibrillation (AF) detection. Approach. Built upon a U-Net architecture, SEResUTer incorporates ResNet modules and Transformer encoders to replace convolution blocks, resulting in improved optimization and encoding capabilities. A novel masking strategy is proposed to handle incomplete expert annotations. The model is trained on the QT database (QTDB) and evaluated on the Lobachevsky University Electrocardiography Database (LUDB) to assess its generalization performance. Additionally, the model’s scope is extended to AF detection using the the China Physiological Signal Challenge 2021 (CPSC2021) and the China Physiological Signal Challenge 2018 (CPSC2018) datasets. Main results. The proposed model surpasses existing traditional and deep learning approaches in ECG waveform delineation on the QTDB. It achieves remarkable average F1 scores of 99.14%, 98.48%, and 98.46% for P wave, QRS wave, and T wave delineation, respectively. Moreover, the model demonstrates exceptional generalization ability on the LUDB, achieving average SE, positive prediction rate, and F1 scores of 99.05%, 94.59%, and 94.62%, respectively. By analyzing RR interval differences and the existence of P waves, our method achieves AF identification with 99.20% accuracy on the CPSC2021 test set and demonstrates strong generalization on CPSC2018 dataset. Significance. The proposed approach enables highly accurate ECG waveform delineation and AF detection, facilitating automated analysis of large-scale ECG recordings and improving the diagnosis of cardiac abnormalities.

Multi-Parametric Delineation Approach for Homogeneous Sectioning of Asphalt Pavements

The demand for preserving existing roadway infrastructure has been increasing to regulate expensive reconstruction activities. The maintenance of homogeneous road sections is one of the approaches to economize the overall management of pavement systems. The existing homogeneous delineation methods consider one or two parameters for segmenting the pavements based on similar characteristics, which are found to be a repetitive process. Also, there is a need to consider multiple parameters that represent the functional, structural, and traffic characteristics in segmentation process. Therefore, the objective of this study was to develop a multi-parameter-based delineation approach (MPDA) to segment the pavements into subsections with similar features considering functional, structural, and traffic characteristics. Deflection bowl parameters, unified pavement health index (functional performance metric), surface layer modulus, and traffic reported in terms of AADT were employed for developing a multi-parametric delineation index (MPDI). A total of 1781 datapoints covering 26 road sections in the State of Andhra Pradesh, India, were used. The C-charts method-based segmentation for MPDI was applied to obtain the homogeneous sections. The devised approach was found to be efficient in segmenting the pavements as well as robust in selecting suitable maintenance strategies for each group of the homogeneous sections. Further, the segmentation processes were automated for easier implementation by the agencies.

Mapping Urban Structure Types Based on Remote Sensing Data—A Universal and Adaptable Framework for Spatial Analyses of Cities

In the face of growing 21st-century urban challenges, this study emphasizes the role of remote sensing data in objectively defining urban structure types (USTs) based on morphology. While numerous UST delineation approaches exist, few are universally applicable due to data constraints or impractical class schemes. This article attempts to tackle this challenge by summarizing important approaches dealing with the computation of USTs and to condense their contributions to the field of research within a single comprehensive framework. Hereby, this framework not only serves as a conjunctive reference for currently existing implementations, but is also independent regarding the input data, spatial scale, or targeted purpose of the mapping. It consists of four major steps: (1) the collection of suitable data sources to describe the building morphology as a key input, (2) the definition of a spatial mapping unit, (3) the parameterization of the mapping units, and (4) the final classification of the mapping units into urban structure types. We outline how these tasks can lead to a UST classification which fits the users’ needs based on their available input data. At the same time, the framework can serve as a protocol for future studies where USTs are mapped, or new approaches are presented. This article closes with an application example for three different cities to underline the flexibility and applicability of the proposed framework while maintaining maximized objectivity and comparability. We recommend this framework as a guideline for the use-specific mapping of USTs and hope to contribute to past and future research on this topic by fostering the implementation of this concept for the spatial analysis and a better understanding of complex urban environments.

Mycobacteriales taxonomy using network analysis-aided, context-uniform phylogenomic approach for non-subjective genus demarcation.

A robust taxonomy is essential for the organized study of prokaryotes and the effective communication of microbial knowledge. The genus rank is the mainstay of biological classification as it brings together under a common name a group of closely related organisms sharing the same recent ancestry and similar characteristics. Despite the unprecedented resolution afforded by whole-genome sequencing in defining evolutionary relationships, a consensus approach for phylogenomics-based prokaryotic genus delineation remains elusive. Taxonomists use different demarcation criteria, sometimes leading to genus rank over-splitting and the creation of multiple new genera. This work reports a simple, reliable, and standardizable method that seeks to minimize subjectivity in genomics-based demarcation of prokaryotic genera, exemplified through application to the order Mycobacteriales. Formal descriptions of proposed taxonomic changes based on our study are included.

Taxonomy, Phylogeny, and Life Cycle of Four Neophysopella Species on Meliosma myriantha and M. oldhamii in Korea

The present study performed an in-depth investigation of rust diseases affecting Meliosma myriantha and Meliosma oldhamii trees (Sabiaceae) in Korea. The analysis identified four distinct species of the genus Neophysopella (Pucciniales) as the causal agents. Among these, N. hornotina was found to infect only M. oldhamii, whereas three Neophysopella species (N. meliosmae, N. meliosmae-myrianthae, and N. vitis) were parasitic on M. myriantha. To our knowledge, this is the first report of the former two species (N. hornotina and N. meliosmae) in Korea. In addition, we specified their alternate host plants for two heteroecious species (N. meliosmae-myrianthae and N. vitis) in Korea, completing the life cycles of the four rust species, and provided detailed morphological descriptions at each stage of their life cycles. Phylogenetic relationships of these rust species were uncovered using a comprehensive sample size, and we have constructed a phylogenetic tree for Neophysopella using the cytochrome c oxidase subunit III (cox3) gene sequences, demonstrating an effective approach for species delineation within this genus. The findings contribute to identifying and managing rust diseases affecting Meliosma species.

Catchment-Area Delineation Approach Considering Travel Purposes for Station-Level Ridership Prediction Task

Station-catchment-area delineation is a key component of direct ridership models for urban rail-transport systems as it can determine the relationship between the urban-rail-transit station-level ridership and the variables within the station catchment area. The neglect of differences in the passenger-flow distribution for different travel purposes in previous studies has led to low accuracy of the obtained walk-to-station distances. Therefore, this paper proposes a station-catchment-area delineation method which is based on web map data to obtain accurate walk-to-station distances and considers differences in the distance thresholds and the ridership attraction intensity (RAI) for six travel purposes (corresponding to commercial, medical, residential, educational, administrative, and recreational land uses). In the case study, the ridership data of Xi’an Metro, the 2015 Xi’an Residential Travel Survey data, and the corresponding Gaode Map data are employed to extract passengers’ walking-distance distribution for several travel purposes to delineate the station catchment areas and build direct ridership models. Several geographically weighted regression (GWR) models are constructed to evaluate and examine the effects of the various station-catchment-area delineation methods on the model findings. The obtained results show that the proposed station-catchment-area delineation method significantly improves the ridership prediction performance compared with the traditional circular-buffer method, with the entry and exit ridership prediction accuracy improving by 3.57% and 6.65% on average, respectively. Finally, this study will guide transportation planners on how to delineate station catchment areas when constructing direct-demand models for urban rail stations.

A deep learning approach for automatic tumor delineation in stereotactic radiotherapy for non-small cell lung cancer using diagnostic PET-CT and planning CT.

Accurate delineation of tumor targets is crucial for stereotactic body radiation therapy (SBRT) for non-small cell lung cancer (NSCLC). This study aims to develop a deep learning-based segmentation approach to accurately and efficiently delineate NSCLC targets using diagnostic PET-CT and SBRT planning CT (pCT). The diagnostic PET was registered to pCT using the transform matrix from registering diagnostic CT to the pCT. We proposed a 3D-UNet-based segmentation method to segment NSCLC tumor targets on dual-modality PET-pCT images. This network contained squeeze-and-excitation and Residual blocks in each convolutional block to perform dynamic channel-wise feature recalibration. Furthermore, up-sampling paths were added to supplement low-resolution features to the model and also to compute the overall loss function. The dice similarity coefficient (DSC), precision, recall, and the average symmetric surface distances were used to assess the performance of the proposed approach on 86 pairs of diagnostic PET and pCT images. The proposed model using dual-modality images was compared with both conventional 3D-UNet architecture and single-modality image input. The average DSC of the proposed model with both PET and pCT images was 0.844, compared to 0.795 and 0.827, when using 3D-UNet and nnUnet. It also outperformed using either pCT or PET alone with the same network, which had DSC of 0.823 and 0.732, respectively. Therefore, our proposed segmentation approach is able to outperform the current 3D-UNet network with diagnostic PET and pCT images. The integration of two image modalities helps improve segmentation accuracy.

The squaretail mullet Ellochelon vaigiensis (Quoy & Gaimard, 1825) a complex of cryptic species?

Abstract The teleost family Mugilidae is speciose with uniform morpho-anatomical characteristics, which render species identification difficult. The dna barcoding technique has, however, proven to be a precise and reliable approach for species delineation. To date, dna barcoding flags numerous polyphyletic species in Mugilidae that probably correspond to species complexes and that call for further taxonomical investigation. Among these species, the squaretail mullet Ellochelon vaigiensis is an interesting case study because, unlike other mullet species, it is easily identified from its unique phenotype. Recent studies of genetic diversity in this monotypic species have revealed two lineages, located either in the Indo-Pacific (Polynesia and Taiwan waters) or along Australian shores. In this study, a third lineage is described in the North of the Indian Ocean, based on nucleotide polymorphisms of Cytochrome c oxidase 1 and barcodes available in bold and Cytochrome b. Despite genetic divergences that exceed the intraspecific threshold, there was no morpho-anatomical difference among specimens of the north Indian Ocean vs. Indo-Pacific or Australia. These molecular results suggest nominal species of Ellochelon vaigiensis within a cryptic species complex.

A deep learning approach for automatic delineation of clinical target volume in stereotactic partial breast irradiation (S-PBI)

Accurate and efficient delineation of the clinical target volume (CTV) is of utmost significance in post-operative breast cancer radiotherapy. However, CTV delineation is challenging as the exact extent of microscopic disease encompassed by CTV is not visualizable in radiological images and remains uncertain. We proposed to mimic physicians’ contouring practice for CTV segmentation in stereotactic partial breast irradiation (S-PBI) where CTV is derived from tumor bed volume (TBV) via a margin expansion followed by correcting the extensions for anatomical barriers of tumor invasion (e.g. skin, chest wall). We proposed a deep-learning model, where CT images and the corresponding TBV masks formed a multi-channel input for a 3D U-Net based architecture. The design guided the model to encode the location-related image features and directed the network to focus on TBV to initiate CTV segmentation. Gradient weighted class activation map (Grad-CAM) visualizations of the model predictions revealed that the extension rules and geometric/anatomical boundaries were learnt during model training to assist the network to limit the expansion to a certain distance from the chest wall and the skin. We retrospectively collected 175 prone CT images from 35 post-operative breast cancer patients who received 5-fraction partial breast irradiation regimen on GammaPod. The 35 patients were randomly split into training (25), validation (5) and test (5) sets. Our model achieved mean (standard deviation) of 0.94 (±0.02), 2.46 (±0.5) mm, and 0.53 (±0.14) mm for Dice similarity coefficient, 95th percentile Hausdorff distance, and average symmetric surface distance respectively on the test set. The results are promising for improving the efficiency and accuracy of CTV delineation during on-line treatment planning procedure.