Diagnostic software proposal for bone scan follow-up using false color based on the gammagrams analysis from gray tone histograms

Abstract

In this work we introduce a technique to speed up the interpretation of bone scans with the aim of determining the presence or absence of metastatic disease. We use gray tone histograms, resembling the use of band-pass filters, in order to ensure a reliable interpretation of the bone scan, therefore providing an accurate diagnosis. We draw particular attention to three cases. The first case corresponds to shifted histograms. If the histogram is shifted toward the origin, the bone scan is free of metastasis. If it is shifted to the right and slightly broadened, this indicates the presence of a bone scan anomaly other than metastasis. On the other hand, if the histogram is broadened and shifted to the left, this suggests the presence of metastatic disease. The second case corresponds to a histogram with noticeable fluctuations, indicating the presence of metastasis. Such fluctuations could become local maxima peaks, indicating the advance of the metastasis. The third case corresponds to the false color results, displayed in terms of the gray tones, observed in the histogram. Such false color is assigned from the construction of a 7-color palette and is selected in terms of the gray tones range. This eases the ad hoc false color assignation for visualization purposes. The final diagnosis is carried out in terms of the color, geometry, extension, and location of the region of interest in the images. Our proposed technique has the potential to be used in high-demand oncology centers due to its simplicity and diagnostic efficiency, confirmed and tested by specialists in the Centro Medico Siglo XXI (XXI Century Medical Center), CDMX-México.

System for medical image visualization: SYFANMI

“…in the image analysis, providing support instead of performing a diagnosis. Our software tool is intended to support the specialist in the decision-making process, to provide a more reliable prognosis, a more accurate treatment, and to improve patient life quality 12 . The system is capable of analyzing the following image types: radiographs, mammography, echography and gammagrams.…”

“…We integrate into a single system or software tool (SYFAN-MI) elements from artificial vision, digital image processing (DIP) 3,13 , in addition to functions/algorithms developed specifically for this proposal, to be applied in the visualization and analysis of medical images, to provide support in the decision-making by the expert in a specific area such as radiology, imaging, and oncology. SYFANMI offers the experts the possibility to process the images based on the object under study, using the available functions, including those devoted to Region of Interest (ROI) segmentation, according to the study or oncologic analysis 12 .…”

“…Analyzing the radiographs of a humeral body, we notice that it is possible to characterize the progress of bone degradation by simply visualizing the images in color 12,13 . Hence, in SYFANMI 1 , the function image transform is available which assigns a false color to a gray-tone image following several criteria.…”

“…This function has been successfully used to detect early bone metastatic disease [11][12][13] in bone scans, since the gray tone image transformation into color allows distinguishing between healthy and infiltrated cases, easing the detection of the anomalies in the image (Fig. 7).…”

“…The user has the freedom to use this function if necessary to identify the changes or bone degradations by means of the graphs or to find some details that were not observed during the image visualization in color due to the tone ranges contained in a single color. This option is particularly useful in bone degradation as in the detection of osteoporosis and verification of a bone anomaly progress, early metastatic disease, and others 10,11,12 . This is the most effective technique for analyzing an image to find bone degradation, even in mild cases, but it is a time-consuming process if the row number is large.…”