Naslov
Estimation of Missing Parameters for DICOM to 8-bit X-ray Image Export

Autori
Hrzic, Franko ; Napravnik, Mateia ; Bazdaric, Robert ; Stajduhar, Ivan ; Mamula, Mihaela ; Miletic, Damir ; Tschauner, Sebastian

Kolaboracija
Department of Radiology, Division of Pediatric Radiology, Medical University of Graz, Graz, Austria

Vrsta, podvrsta i kategorija rada
Radovi u zbornicima skupova, cjeloviti rad (in extenso), znanstveni

Izvornik
Proc. of the International Conference on Electrical, Computer, Communications and Mechatronics Engineering (ICECCME) / - : Institute of Electrical and Electronics Engineers (IEEE), 2022, 1-6

ISBN
978-1-6654-7095-7

Skup
International Conference on Electrical, Computer, Communications and Mechatronics Engineering (ICECCME 2022)

Mjesto i datum
Malé, Maldivi, 16.11.2022. - 18.11.2022

Vrsta sudjelovanja
Predavanje

Vrsta recenzije
Međunarodna recenzija

Ključne riječi
X-ray ; Entropy ; DICOM export ; Medical imaging

Sažetak
Digital Imaging and Communication in Medicine (DICOM) is a standard data storage format for many data types, including X-ray images. X-ray images stored in DICOM format commonly have 12 to 16 bits depth. To account for the limitations of computers' displays and the receptive properties of a human eye, the images are often adapted. The adapted format commonly has only 8-bit depth, resulting in information loss during the conversion. To preserve as much information as possible and display enhanced information that is useful to medical experts, the raw data is transformed into displayable data by focusing only on an area of interest in raw values. A window-shaped area defines the area of interest with two important parameters: windows level and width. If the aforementioned parameters are missing, the conversion cannot be correctly performed. We propose a method that successfully calculates missing parameters with no statistically significant difference. The method is evaluated on 933 DICOM files of X-ray images representing various body parts collected from two hospitals. Furthermore, we have investigated how much information is preserved during the conversion. The research showed that some methods can have higher entropy (amount of information) than the commonly used method based on window level and width parameters, concerning the raw data stored in DICOM.

Izvorni jezik
Engleski

Znanstvena područja
Računarstvo, Temeljne tehničke znanosti, Interdisciplinarne tehničke znanosti, Temeljne medicinske znanosti

POVEZANOST RADA