AimComputer-aided diagnosis (CAD) software for bone scintigrams have recently been introduced as a clinical quality assurance tool. The purpose of this study was to compare the diagnostic accuracy of two CAD systems, one based on a European and one on a Japanese training database, in a group of bone scans from Japanese patients.MethodThe two CAD software are trained to interpret bone scans using training databases consisting of bone scans with the desired interpretation, metastatic disease or not. One software was trained using 795 bone scans from European patients and the other with 904 bone scans from Japanese patients. The two CAD softwares were evaluated using the same group of 257 Japanese patients, who underwent bone scintigraphy because of suspected metastases of malignant tumors in 2009. The final diagnostic results made by clinicians were used as gold standard.ResultsThe Japanese CAD software showed a higher specificity and accuracy compared to the European CAD software [81 vs. 57 % (p < 0.05) and 82 vs. 61 % (p < 0.05), respectively]. The sensitivity was 90 % for the Japanese CAD software and 83 % for the European CAD software (n.s).ConclusionThe CAD software trained with a Japanese database showed significantly higher performance than the corresponding CAD software trained with a European database for the analysis of bone scans from Japanese patients. These results could at least partly be caused by the physical differences between Japanese and European patients resulting in less influence of attenuation in Japanese patients and possible different judgement of count intensities of hot spots.

Automated segmentation of the skeleton in a Japanese patient group was more successful when the CAD system based on a Japanese atlas was used than when the corresponding system based on a European atlas was used. The results of this study indicate that it is of value to use a skeletal atlas based on normal Japanese bone scans in a CAD system for Japanese patients.

In recent years, the application of artificial intelligence (AI) to medical images has advanced rapidly. Especially since the learning method called deep learning has spread, we have made remarkable progress including precision. It is application ranges from detection of abnormal lesion on the image to screening assistants or selection of drugs necessary for treatment. Study of nuclear medicine is progressing with AI, and reports are also made on bone scintigraphy and PET images. In nuclear cardiology, software for detecting abnormal lesion of myocardium using artificial neural network (ANN) by machine learning and making it useful for image interpretation aid has been developed. In the future, it is expected that study and development will also advance interpretation assistance by deep learning and abnormal lesion detection. However, for deep learning, the number of images required for study becomes enormous and it is a point of discussion.

In this paper, we propose new diagnostic assist systems of medical images using deep learning algorithms. Specifically, we aim to develop a diagnostic support system for the very early stage of chronic obstructive pulmonary disease (COPD) based on the CT images. It is said that COPD is a disease that develops due to long-term smoking, and it is said that there are a large number of latent onset reserve forces. By discovering this COPD in the very early period 0 and improving the living conditions, subsequent severity can be avoided in many cases, so a system that will help diagnosis by professional radiologists is needed. We show the some experimental results examined by the constructed system.

Background: Digital anthropomorphic phantoms have gradually gained an important role in nuclear medicine imaging. The aim of this study was to generate digital phantom models simulated with cardiac and respiratory motions and to evaluate motion-induced artifact using a quantitative software program developed with artificial intelligence (AI) technology in myocardial perfusion single-photon emission computed tomography (SPECT)