| Περίληψη: | Medical information, composed of clinical data, images and other physiological signals, has become an essential part of a patient’s care, whether during screening, the diagnostic stage or the treatment phase. Data in the form of images and signals form part of each patient’s medical file, and as such have to be stored and often transmitted from one place to another. All this huge volume of information must be stored in medical records. It has been estimated that the total amount of information to be stored in a typical hospital database increases at the amount of data of over 1015 bytes annually. Even with the biggest magnetic and optical disks and other database hardware, storage space must be conserved somehow. The most effective method of storage space saving is data compression, so we must consider methods used for compressing images.
The current study presents a quantitative approach towards visually lossless compression ratio (CR) threshold determination in digitized mammograms. This is achieved by identifying quantitative image quality metrics that reflect radiologists’ visual perception in distinguishing between original and wavelet-compressed mammographic regions of interest containing microcalcification clusters (MCs) and normal parenchyma, originating from images from the Digital Database for Screening Mammography. The image quality of wavelet compressed mammograms is evaluated quantitatively by means of eight image quality metrics of different computational principles and qualitatively by three radiologists employing a five-point rating scale. The accuracy of the objective metrics is investigated in terms of (1) their correlation (r) with qualitative assessment and (2) ROC analysis (Az index), employing pooled radiologists’ rating scores as ground truth. The quantitative metrics mean square error, mean absolute error, peak signal-to-noise ratio, and structural similarity demonstrated strong correlation with pooled radiologists’ ratings and the highest area under ROC curve. For each quantitative metric, the highest accuracy values of corresponding ROC curves were used to define metric cut-off values. The metrics cut-off values were subsequently used to suggest a visually lossless CR threshold, estimated to be between 25:1 and 40:1 for the dataset analyzed. Results indicate the potential of the quantitative metrics approach in predicting visually lossless CRs in case of MCs in mammography.
|
|---|
