Zhigang Liang

Multilevel binomial logistic prediction model for malignant pulmonary nodules based on texture features of CT image

PURPOSE To introduce multilevel binomial logistic prediction model-based computer-aided diagnostic (CAD) method of small solitary pulmonary nodules (SPNs) diagnosis by combining patient and image characteristics by textural features of CT image. MATERIALS AND METHODS Describe fourteen gray level co-occurrence matrix textural features obtained from 2171 benign and malignant small solitary pulmonary nodules, which belongs to 185 patients. Multilevel binomial logistic model is applied to gain these initial insights. RESULTS Five texture features, including Inertia, Entropy, Correlation, Difference-mean, Sum-Entropy, and age of patients own aggregating character on patient-level, which are statistically different (P<0.05) between benign and malignant small solitary pulmonary nodules. CONCLUSION Some gray level co-occurrence matrix textural features are efficiently descriptive features of CT image of small solitary pulmonary nodules, which can profit diagnosis of earlier period lung cancer if combined patient-level characteristics to some extent.

Effects of different compression techniques on diagnostic accuracies of breast masses on digitized mammograms

Background: The JPEG 2000 compression technique has recently been introduced into the medical imaging field. It is critical to understand the effects of this technique on the detection of breast masses on digitized images by human observers. Purpose: To evaluate whether lossless and lossy techniques affect the diagnostic results of malignant and benign breast masses on digitized mammograms. Material and Methods: A total of 90 screen-film mammograms including craniocaudal and lateral views obtained from 45 patients were selected by two non-observing radiologists. Of these, 22 cases were benign lesions and 23 cases were malignant. The mammographic films were digitized by a laser film digitizer, and compressed to three levels (lossless and lossy 20:1 and 40:1) using the JPEG 2000 wavelet-based image compression algorithm. Four radiologists with 10–12 years’ experience in mammography interpreted the original and compressed images. The time interval was 3 weeks for each reading session. A five-point malignancy scale was used, with a score of 1 corresponding to definitely not a malignant mass, a score of 2 referring to not a malignant mass, a score of 3 meaning possibly a malignant mass, a score of 4 being probably a malignant mass, and a score of 5 interpreted as definitely a malignant mass. The radiologists’ performance was evaluated using receiver operating characteristic analysis. Results: The average Az values for all radiologists decreased from 0.8933 for the original uncompressed images to 0.8299 for the images compressed at 40:1. This difference was not statistically significant. The detection accuracy of the original images was better than that of the compressed images, and the Az values decreased with increasing compression ratio. Conclusion: Digitized mammograms compressed at 40:1 could be used to substitute original images in the diagnosis of breast cancer.

Comparison of paroxetine and agomelatine in depressed type 2 diabetes mellitus patients: a double-blind, randomized, clinical trial

Background Comorbid depression/anxiety in type 2 diabetes mellitus (DM) patients is highly prevalent, affecting both diabetes control and quality of life. However, the best treating method for depression/anxiety in type 2 DM patients is still unclear. This study was conducted to compare the efficacy of paroxetine and agomelatine on depression/anxiety and metabolic control of type 2 DM patients. Methods A total of 116 depressed, type 2 DM patients were recruited for 12 weeks treatment. Patients were randomly assigned to receive either paroxetine or agomelatine. Hamilton Depression Rating Scale and Hamilton Anxiety Rating Scale were used to assess depression and anxiety, respectively. Hemoglobin A1c, fasting plasma glucose, and body mass index were assessed at baseline and at the end of the trial. Results At the end of the trial, there were 34 (60.7%) responders and 22 (39.3%) remissions in paroxetine group; and 38 (63.3%) responders and 26 (43.3%) remissions in agomelatine group. Compared to paroxetine group, lower depression scores were observed in agomelatine group. Fasting plasma glucose and body mass index were not significantly different after 12 weeks treatment between the two groups, but agomelatine group had a significantly lower final hemoglobin A1c level compared to paroxetine group. The two antidepressants had comparable acceptability. Conclusion These results showed that compared to paroxetine, agomelatine might have some advantages in treating symptoms of depression/anxiety and glycemic control in depressed type 2 DM patients. The clinical applicability of agomelatine shows greater promise and should be explored further. Limited by the relatively small samples, future studies are needed to verify and support our findings.

Comparison of dry laser printer versus paper printer in full-field digital mammography.

Background: Paper printers have been used to document radiological findings in some hospitals. It is critical to establish whether paper printers can achieve the same efficacy and quality as dry laser printers for full-field digital mammography (FFDM). Purpose: To compare the image quality and detection rate of dry laser printers and paper printers for FFDM. Material and Methods: Fifty-five cases (25 with single clustered microcalcifications and 30 controls) were selected by a radiologist not participating in the image review. All images were printed on film and paper by one experienced mammography technologist using the processing algorithm routinely used for our mammograms. Two radiologists evaluated hard copies from dry laser printers and paper printers for image quality and detectability of clustered microcalcifications. For the image quality comparisons, agreement between the reviewers was evaluated by means of kappa statistics. The significance of differences between both of the printers was determined using Wilcoxons signed-rank test. The detection rate of two printing systems was evaluated using receiver operating characteristic (ROC) analysis. Results: From 110 scores (55 patients, two readers) per printer system, the following quality results were achieved for dry laser printer images: 70 (63.6%) were rated as good and 40 (36.4%) as moderate. By contrast, for the paper printer images, 25 scores (22.7%) were rated as good and 85 (77.3%) as moderate. Therefore, the image quality of the dry laser printer was superior to that achieved by the paper printer (P=0.00). The average area-under-the-curve (Az) values for the dry laser printer and the paper printer were 0.991 and 0.805, respectively. The difference was 0.186. Results of ROC analysis showed significant difference in observer performance between the dry laser printer and paper printer (P=0.0015). Conclusion: The performance of dry laser printers is superior to that of paper printers. Paper printers should not be used in FFDM.

Comparison of diagnostic accuracy of breast masses using digitized images versus screen-film mammography

Background: Medical film digitizers play an important transitory role as digital–analogue bridges in radiology. Digitized mammograms require evaluation of performance to assure medical image quality. Purpose: To compare the diagnostic accuracy in the interpretation of breast masses using original screen-film mammograms versus digitized images. Material and Methods: A total of 72 female patients between 55 and 81 years of age suspected of having breast cancer were selected by two non-observing radiologists. Of these, 31 cases were benign lesions and 41 cases were cancer. The mammography films were digitized using a laser film digitizer. Three radiologists, each with more than 10 years of experience in mammography, interpreted the screen-film mammograms and digitized images respectively. The time interval was 4 weeks. A four-point malignancy scale was used, with 1 defined as definitely not malignant, 2 as probably not malignant, 3 as probably malignant, and 4 as definitely malignant. Receiver operating characteristic (ROC) curves, sensitivity, and specificity were compared. Results: The average area-under-the-curve (Az) value of the original screen-film mammograms was 0.921, and the average Az value of the digitized images was 0.859. This difference was not statistically significant (P=0.131). The detection specificity of extremely dense breasts was lower than that for other breast compositions for both digitized images and screen-film mammograms. No statistical significance in sensitivity and specificity was observed between digitized images and mammograms for each breast composition. Original screen-film mammograms were observed to perform better than digitized images. Conclusion: Digitized images with a spatial resolution of 175 µm can be used instead of screen-film mammograms in the diagnosis of breast cancer.

Contourlet textual features: improving the diagnosis of solitary pulmonary nodules in two dimensional CT images.

Objective To determine the value of contourlet textural features obtained from solitary pulmonary nodules in two dimensional CT images used in diagnoses of lung cancer. Materials and Methods A total of 6,299 CT images were acquired from 336 patients, with 1,454 benign pulmonary nodule images from 84 patients (50 male, 34 female) and 4,845 malignant from 252 patients (150 male, 102 female). Further to this, nineteen patient information categories, which included seven demographic parameters and twelve morphological features, were also collected. A contourlet was used to extract fourteen types of textural features. These were then used to establish three support vector machine models. One comprised a database constructed of nineteen collected patient information categories, another included contourlet textural features and the third one contained both sets of information. Ten-fold cross-validation was used to evaluate the diagnosis results for the three databases, with sensitivity, specificity, accuracy, the area under the curve (AUC), precision, Youden index, and F-measure were used as the assessment criteria. In addition, the synthetic minority over-sampling technique (SMOTE) was used to preprocess the unbalanced data. Results Using a database containing textural features and patient information, sensitivity, specificity, accuracy, AUC, precision, Youden index, and F-measure were: 0.95, 0.71, 0.89, 0.89, 0.92, 0.66, and 0.93 respectively. These results were higher than results derived using the database without textural features (0.82, 0.47, 0.74, 0.67, 0.84, 0.29, and 0.83 respectively) as well as the database comprising only textural features (0.81, 0.64, 0.67, 0.72, 0.88, 0.44, and 0.85 respectively). Using the SMOTE as a pre-processing procedure, new balanced database generated, including observations of 5,816 benign ROIs and 5,815 malignant ROIs, and accuracy was 0.93. Conclusion Our results indicate that the combined contourlet textural features of solitary pulmonary nodules in CT images with patient profile information could potentially improve the diagnosis of lung cancer.

Intra-tumoural heterogeneity characterization through texture and colour analysis for differentiation of non-small cell lung carcinoma subtypes

Radiomics has shown potential in disease diagnosis, but its feasibility for non-small cell lung carcinoma (NSCLC) subtype classification is unclear. This study aims to explore the diagnosis value of texture and colour features from positron emission tomography computed tomography (PET-CT) images in differentiation of NSCLC subtypes: adenocarcinoma (ADC) and squamous cell carcinoma (SqCC). Two patient cohorts were retrospectively collected into a dataset of 341 18F-labeled 2-deoxy-2fluoro-d-glucose ([18F] FDG) PET-CT images of NSCLC tumours (125 ADC, 174 SqCC, and 42 cases with unknown subtype). Quantification of texture and colour features was performed using freehand regions of interest. The relation between extracted features and commonly used parameters such as age, gender, tumour size, and standard uptake value (SUVmax) was explored. To classify NSCLC subtypes, support vector machine algorithm was applied on these features and the classification performance was evaluated by receiver operating characteristic curve analysis. There was a significant difference between ADC and SqCC subtypes in texture and colour features (P  <  0.05); this showed that imaging features were significantly correlated to both SUVmax and tumour diameter (P  <  0.05). When evaluating classification performance, features combining texture and colour showed an AUC of 0.89 (95% CI, 0.78-1.00), colour features showed an AUC of 0.85 (95% CI, 0.71-0.99), and texture features showed an AUC of 0.68 (95% CI, 0.48-0.88). DeLongs test showed that AUC was higher for features combining texture and colour than that for texture features only (P  =  0.010), but not significantly different from that for colour features only (P  =  0.328). HSV colour features showed a similar performance to RGB colour features (P  =  0.473). The colour features are promising in the refinement of NSCLC subtype differentiation, and features combining texture and colour of PET-CT images could result in better classification performance.

Contourlet-based hippocampal magnetic resonance imaging texture features for multivariant classification and prediction of Alzheimer’s disease

The study is aimed to assess whether the addition of contourlet-based hippocampal magnetic resonance imaging (MRI) texture features to multivariant models improves the classification of Alzheimer’s disease (AD) and the prediction of mild cognitive impairment (MCI) conversion, and to evaluate whether Gaussian process (GP) and partial least squares (PLS) are feasible in developing multivariant models in this context. Clinical and MRI data of 58 patients with probable AD, 147 with MCI, and 94 normal controls (NCs) were collected. Baseline contourlet-based hippocampal MRI texture features, medical histories, symptoms, neuropsychological tests, volume-based morphometric (VBM) parameters based on MRI, and regional CMgl measurement based on fluorine-18 fluorodeoxyglucose-positron emission tomography were included to develop GP and PLS models to classify different groups of subjects. GPR1 model, which incorporated MRI texture features and was based on GPG, performed better in classifying different groups of subjects than GPR2 model, which used the same algorithm and had the same data as GPR1 except that MRI texture features were excluded. PLS model, which included the same variables as GPR1 but was based on the PLS algorithm, performed best among the three models. GPR1 accurately predicted 82.2% (51/62) of MCI convertors confirmed during the 2-year follow-up period, while this figure was 53 (85.5%) for PLS model. GPR1 and PLS models accurately predicted 58 (79.5%) vs. 61 (83.6%) of 73 patients with stable MCI, respectively. For seven patients with MCI who converted to NCs, PLS model accurately predicted all cases (100%), while GPR1 predicted six (85.7%) cases. The addition of contourlet-based MRI texture features to multivariant models can effectively improve the classification of AD and the prediction of MCI conversion to AD. Both GPR and LPS models performed well in the classification and predictive process, with the latter having significantly higher classification and predictive accuracies. Advances in knowledge: We combined contourlet-based hippocampal MRI texture features, medical histories, symptoms, neuropsychological tests, volume-based morphometric (VBM) parameters, and regional CMgl measurement to develop models using GP and PLS algorithms to classify AD patients.

Clinical evaluation of whole-body oncologic PET with time-of-flight and point-spread function for the hybrid PET/MR system

PURPOSE Hybrid positron emission tomography/magnetic resonance (PET/MR) imaging is a new multimodality imaging technology that can provide structural and functional information simultaneously. The aim of this study was to investigate the effects of the time-of-flight (TOF) and point-spread function (PSF) on small lesions observed in PET/MR images from clinical patient image sets. MATERIALS AND METHODS This study evaluated 54 small lesions in 14 patients who had undergone 18F-fluorodeoxyglucose (FDG) PET/MR. Lesions up to 30mm in diameter were included. The PET data were reconstructed with a baseline ordered-subsets expectation-maximization (OSEM) algorithm, OSEM+PSF, OSEM+TOF and OSEM+TOF+PSF. PET image quality and small lesions were visually evaluated and scored by a 3-point scale. A quantitative analysis was then performed using the mean and maximum standardized uptake value (SUV) of the small lesions (SUVmean and SUVmax). The lesions were divided into two groups according to the long-axis diameter and the location respectively and evaluated with each reconstruction algorithm. We also evaluated the background signal by analyzing the SUVliver. RESULTS OSEM+TOF+PSF provided the highest value and OSEM+TOF or PSF showed a higher value than OSEM for the visual assessment and quantitative analysis. The combination of TOF and PSF increased the SUVmean by 26.6% and the SUVmax by 30.0%. The SUVliverwas not influenced by PSF or TOF. For the OSEM+TOF+PSF model, the change in SUVmean and SUVmax for lesions <10mm in diameter was 31.9% and 35.8%, and 24.5% and 27.6% for lesions 10-30mm in diameter, respectively. The abdominal lesions obtained the higher SUV than those of chest on the images with TOF and/or PSF. CONCLUSION Application of TOF and PSF significantly increased the SUV of small lesions in hybrid PET/MR images, potentially improving small lesion detectability.