Dataset Information

DP2: Distributed 3D image segmentation using micro-labor workforce.

ABSTRACT:

Summary

This application note describes a new scalable semi-automatic approach, the Dual Point Decision Process, for segmentation of 3D structures contained in 3D microscopy. The segmentation problem is distributed to many individual workers such that each receives only simple questions regarding whether two points in an image are placed on the same object. A large pool of micro-labor workers available through Amazon's Mechanical Turk system provides the labor in a scalable manner.

Availability and implementation

Python-based code for non-commercial use and test data are available in the source archive at https://sites.google.com/site/imagecrowdseg/.

Contact

rgiuly@ucsd.edu

Supplementary information

Supplementary data are available at Bioinformatics online.

SUBMITTER: Giuly RJ

PROVIDER: S-EPMC3654713 | biostudies-literature | 2013 May

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

DP2: Distributed 3D image segmentation using micro-labor workforce.

Giuly Richard J RJ Kim Keun-Young KY Ellisman Mark H MH

Bioinformatics (Oxford, England) 20130410 10

<h4>Summary</h4>This application note describes a new scalable semi-automatic approach, the Dual Point Decision Process, for segmentation of 3D structures contained in 3D microscopy. The segmentation problem is distributed to many individual workers such that each receives only simple questions regarding whether two points in an image are placed on the same object. A large pool of micro-labor workers available through Amazon's Mechanical Turk system provides the labor in a scalable manner.<h4>Av ...[more]

PMID: 23574738

Similar Datasets

Project description:BackgroundHigh quality 3D information of the microscopic plant tissue morphology-the spatial organization of cells and intercellular spaces in tissues-helps in understanding physiological processes in a wide variety of plants and tissues. X-ray micro-CT is a valuable tool that is becoming increasingly available in plant research to obtain 3D microstructural information of the intercellular pore space and individual pore sizes and shapes of tissues. However, individual cell morphology is difficult to retrieve from micro-CT as cells cannot be segmented properly due to negligible density differences at cell-to-cell interfaces. To address this, deep learning-based models were trained and tested to segment individual cells using X-ray micro-CT images of parenchyma tissue samples from apple and pear fruit with different cell and porosity characteristics.ResultsThe best segmentation model achieved an Aggregated Jaccard Index (AJI) of 0.86 and 0.73 for apple and pear tissue, respectively, which is an improvement over the current benchmark method that achieved AJIs of 0.73 and 0.67. Furthermore, the neural network was able to detect other plant tissue structures such as vascular bundles and stone cell clusters (brachysclereids), of which the latter were shown to strongly influence the spatial organization of pear cells. Based on the AJIs, apple tissue was found to be easier to segment, as the porosity and specific surface area of the pore space are higher and lower, respectively, compared to pear tissue. Moreover, samples with lower pore network connectivity, proved very difficult to segment.ConclusionsThe proposed method can be used to automatically quantify 3D cell morphology of plant tissue from micro-CT instead of opting for laborious manual annotations or less accurate segmentation approaches. In case fruit tissue porosity or pore network connectivity is too low or the specific surface area of the pore space too high, native X-ray micro-CT is unable to provide proper marker points of cell outlines, and one should rely on more elaborate contrast-enhancing scan protocols.

Project description:BackgroundEnsuring healthcare delivery is dependent both on the prediction of the future demand for healthcare services and on the estimation and planning for the Health Human Resources needed to properly deliver these services. Although the Health Human Resources planning is a fascinating and widely researched topic, and despite the number of methodologies that have been used, no consensus on the best way of planning the future workforce requirements has been reported in the literature. This paper aims to contribute to the extension and diversity of the range of available methods to forecast the demand for Health Human Resources and assist in tackling the challenge of translating healthcare services to workforce requirements.MethodsA method to empirically quantify the relation between healthcare services and Health Human Resources requirements is proposed. For each one of the three groups of specialties identified-Surgical specialties, Medical specialties and Diagnostic specialties (e.g., pathologists)-a Labor Requirements Function relating the number of physicians with a set of specialty-specific workload and capital variables is developed. This approach, which assumes that health managers and decision-makers control the labor levels more easily than they control the amount of healthcare services demanded, is then applied to a panel dataset comprising information on 142 public hospitals, during a 12-year period.ResultsThis method provides interesting insights on healthcare services delivery: the number of physicians required to meet expected variations in the demand for healthcare, the effect of the technological progress on healthcare services delivery, the time spent on each type of care, the impact of Human Resources concentration on productivity, and the possible resource allocations given the opportunity cost of the physicians' labor.ConclusionsThe empirical method proposed is simple and flexible and produces statistically strong models to estimate Health Human Resources requirements. Moreover, it can enable a more informed allocation of the available resources and help to achieve a more efficient delivery of healthcare services.

Dataset Information

DP2: Distributed 3D image segmentation using micro-labor workforce.

Summary

Availability and implementation

Contact

Supplementary information

Publications

DP2: Distributed 3D image segmentation using micro-labor workforce.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets