Role of ventilation scintigraphy in diagnosis of acute pulmonary embolism: an evaluation using artificial neural networks

Abstract

The purpose of this study was to assess the value of the ventilation study in the diagnosis of acute pulmonary embolism using a new automated method. Either perfusion scintigrams alone or two different combinations of ventilation/perfusion scintigrams were used as the only source of information regarding pulmonary embolism. A completely automated method based on computerised image processing and artificial neural networks was used for the interpretation. Three artificial neural networks were trained for the diagnosis of pulmonary embolism. Each network was trained with 18 automatically obtained features. Three different sets of features originating from three sets of scintigrams were used. One network was trained using features obtained from each set of perfusion scintigrams, including six projections. The second network was trained using features from each set of (joint) ventilation and perfusion studies in six projections. A third network was trained using features from the perfusion study in six projections combined with a single ventilation image from the posterior view. A total of 1,087 scintigrams from patients with suspected pulmonary embolism were used for network training. The test group consisted of 102 patients who had undergone both scintigraphy and pulmonary angiography. Performances in the test group were measured as area under the receiver operation characteristic curve. The performance of the neural network in interpreting perfusion scintigrams alone was 0.79 (95% confidence limits 0.71-0.86). When one ventilation image (posterior view) was added to the perfusion study, the performance was 0.84 (0.77-0.90). This increase was statistically significant ( P=0.022). The performance increased to 0.87 (0.81-0.93) when all perfusion and ventilation images were used, and the increase in performance from 0.79 to 0.87 was also statistically significant ( P=0.016). The automated method presented here for the interpretation of lung scintigrams shows a significant increase in performance when one or all ventilation images are added to the six perfusion images. Thus, the ventilation study has a significant role in the diagnosis of acute lung embolism.

Neural hypernetwork approach for pulmonary embolism diagnosis

“…These markers maintain a predictive role also among patients showing normal blood pressure and wiothout signs of shock: RVD, defined as the presence of increased troponin values or echocardiographic evidence of right ventricle dilatation or increased pulmonary pressure, has been associated with higher mortality [ 24 ] in all patient subsets. All the automatic methods proposed in literature [ 4 – 7 ] for PE diagnosis are in line with the guidelines [ 1 ] in proposing the use of imaging analysis, e.g., CT-angiography, as input space. The risk associated to CT-angiography is still under discussion [ 25 ], and its associated risks can not be neglected.…”

“…The performance of their network, expressed in terms of area under curve of the ROC, encouraged the use of ANN for PE diagnosis [ 6 ]. The performance of this ROC has been improved in a subsequent work, in which the authors trained three ANNs: the first network was trained using features obtained from each set of perfusion scintigrams, the second network was trained using features from each set of (joint) ventilation and perfusion studies in six projections, and the third network was trained using features from the perfusion study in six projections combined with a single ventilation image from the posterior view [ 7 ]. For a comparative analysis of the several models adopted for predicting the presence of acute pulmonary embolism a reading of [ 8 , 9 ] is suggested.…”

Application of Artificial Neural Networks to Identify Alzheimer’s Disease Using Cerebral Perfusion SPECT Data

“…Diagnostic decision support, analysis and interpretation of images are the medical issues solved by means of artificial intelligence methods, the ANN in particular [14,15,16,17,18,19]. Artificial neural networks are used in contemporary nuclear medicine to support the diagnostics of ischemic heart disease [20,21,22], pulmonary embolism [23,24], parathyroid adenoma [24], Alzheimer’s disease [25,26,27,28,29,30] and breast cancer [31]. Apart from those, artificial neural networks are useful tools in dentistry for predicting the number of follow-up visit [32].…”

A study on the value of computer-assisted assessment for SPECT/CT-scans in sentinel lymph node diagnostics of penile cancer as well as clinical reliability and morbidity of this procedure

“…In image-related medical disciplines, this has prompted the evolution of special image processing software, developed to help the medical expert assess large amounts of digital image data in less time to achieve better diagnostic results. The literature reports studies on software applications and analyses of lung scintigraphy for embolism diagnostics [16], and for automated assessment of fluor-deoxyglucose positron-emission-computed-tomography/computertomography (FDG-PET/CT)-examinations in non-small-cell bronchial carcinoma [17]. Moreover, programs for post processing in nuclear medical imaging are now commercially available.…”