An automatic document processing system for medical data extraction

Adamo, Francesco; Attivissimo, Filippo; Nisio, Attilio Di; Spadavecchia, Maurizio

doi:10.1016/j.measurement.2014.10.032

Cited by 20 publications

(11 citation statements)

References 36 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Recently, this technology has improved significantly with improved image processing techniques and the advent of neural networks. In the literature, studies that have specifically used OCR technology for medical data exaction tasks mostly focus on scanned reports for clinical laboratory tests, with reasonably high accuracy ( 19 – 21 ). Adamo et al utilized Tesseract OCR (the same OCR platform as used in our script) to achieve an accuracy of 95% in their extraction system ( 19 ).…”

Section: Discussionmentioning

confidence: 99%

Development and Validation of Automated Visual Field Report Extraction Platform Using Computer Vision Tools

Saifee

Liu

et al. 2021

Front. Med.

View full text Add to dashboard Cite

Purpose: To introduce and validate hvf_extraction_script, an open-source software script for the automated extraction and structuring of metadata, value plot data, and percentile plot data from Humphrey visual field (HVF) report images.Methods: Validation was performed on 90 HVF reports over three different report layouts, including a total of 1,530 metadata fields, 15,536 value plot data points, and 10,210 percentile data points, between the computer script and four human extractors, compared against DICOM reference data. Computer extraction and human extraction were compared on extraction time as well as accuracy of extraction for metadata, value plot data, and percentile plot data.Results: Computer extraction required 4.9-8.9 s per report, compared to the 6.5-19 min required by human extractors, representing a more than 40-fold difference in extraction speed. Computer metadata extraction error rate varied from an aggregate 1.2-3.5%, compared to 0.2-9.2% for human metadata extraction across all layouts. Computer value data point extraction had an aggregate error rate of 0.9% for version 1, <0.01% in version 2, and 0.15% in version 3, compared to 0.8-9.2% aggregate error rate for human extraction. Computer percentile data point extraction similarly had very low error rates, with no errors occurring in version 1 and 2, and 0.06% error rate in version 3, compared to 0.06-12.2% error rate for human extraction.Conclusions: This study introduces and validates hvf_extraction_script, an open-source tool for fast, accurate, automated data extraction of HVF reports to facilitate analysis of large-volume HVF datasets, and demonstrates the value of image processing tools in facilitating faster and cheaper large-volume data extraction in research settings.

show abstract

Section: Discussionmentioning

confidence: 99%

Development and Validation of Automated Visual Field Report Extraction Platform Using Computer Vision Tools

Saifee

Liu

et al. 2021

Front. Med.

View full text Add to dashboard Cite

show abstract

“…The ability to evaluate the online visibility of domain-specific software and to compare online software descriptions with the corresponding (or related) publications is of practical relevance to website indexing and recommendation, namely domain-specific or vertical engines (Kassing, Oosterman, Bozzon, & Houben, 2015;Ms, Kumar, & Mukesh, n.d.). In essence, the problem falls into the research scope of semantic information retrieval, which has been boosting the emergence of alternative search strategies based on domain-specific semantics to improve the accuracy of the retrieved results (Adamo, Attivissimo, Di Nisio, & Spadavecchia, 2015;Gopalakrishnan, Sengottuvelan, Bharathi, & Lokeshkumar, 2018), as well as to determine the similarity between retrieved information (Jiang, Bai, Zhang, & Hu, 2017;Yan, Liu, Wang, Zhang, & Zheng, 2017;Zhang, Wang, Wang, Bi, & Chen, 2014).…”

Section: Related Workmentioning

confidence: 99%

Online visibility of software-related web sites: The case of biomedical text mining tools

Pérez-Rodríguez

Pérez-Pérez

Fdez-Riverola

et al. 2019

Information Processing & Management

View full text Add to dashboard Cite

“…Non-contact and non-destructive imaging techniques, based on properly developed object segmentation and detection algorithms, have been proved to be a viable alternative to contact measurement and diagnostic techniques in a large variety of sectors, ranging from industrial quality control to characterization of devices, to medical imaging and clinical applications. Hence, it is foreseeable the development of automatic vision methods for identification, counting, and measurement of the sea [11,12,13,14,15,16,17,18,19,20].…”

Section: Introductionmentioning

confidence: 99%

Towards Non-Invasive Methods to Assess Population Structure and Biomass in Vulnerable Sea Pen Fields

Chimienti

Nisio

Lanzolla

et al. 2019

Sensors

Self Cite

View full text Add to dashboard Cite

Colonies of the endangered red sea pen Pennatula rubra (Cnidaria: Pennatulacea) sampled by trawling in the northwestern Mediterranean Sea were analyzed. Biometric parameters, such as total length, peduncle length, number of polyp leaves, fresh weight, and dry weight, were measured and related to each other by means of regression analysis. Ad hoc models for future inferencing of colonies size and biomass through visual techniques were individuated in order to allow a non-invasive study of the population structure and dynamics of P. rubra.

show abstract

An automatic document processing system for medical data extraction

Cited by 20 publications

References 36 publications

Development and Validation of Automated Visual Field Report Extraction Platform Using Computer Vision Tools

Development and Validation of Automated Visual Field Report Extraction Platform Using Computer Vision Tools

Online visibility of software-related web sites: The case of biomedical text mining tools

Towards Non-Invasive Methods to Assess Population Structure and Biomass in Vulnerable Sea Pen Fields

Contact Info

Product

Resources

About