Automated Detection of Poor-Quality Scintigraphic Images Using Machine Learning.

Objective In the present study, we have used machine learning algorithm to accomplish the task of automated detection of poor-quality scintigraphic images. We have validated the accuracy of our machine learning algorithm on 99m Tc-methyl diphosphonate ( 99m Tc-MDP) bone scan images. Materials and Methods Ninety-nine patients underwent 99mTC-MDP bone scan acquisition twice at two different acquisition speeds, one at low speed and another at double the speed of the first scan, with patient lying in the same position on the scan table. The low-speed acquisition resulted in good-quality images and the high-speed acquisition resulted in poor-quality images. The principal component analysis (PCA) of all the images was performed and the first 32 principal components (PCs) were retained as feature vectors of the image. These 32 feature vectors of each image were used for the classification of images into poor or good quality using machine learning algorithm (multivariate adaptive regression splines [MARS]). The data were split into two sets, that is, training set and test set in the ratio of 60:40. Hyperparameter tuning of the model was done in which five-fold cross-validation was performed. Receiver operator characteristic (ROC) analysis was used to select the optimal model using the largest value of area under the ROC curve. Sensitivity, specificity, and accuracy for the classification of poor- and good-quality images were taken as metrics for the performance of the algorithm. Result Accuracy, sensitivity, and specificity of the model in classifying poor-quality and good-quality images were 93.22, 93.22, and 93.22%, respectively, for the training dataset and 86.88, 80, and 93.7%, respectively, for the test dataset. Conclusion Machine learning algorithms can be used to classify poor- and good-quality images with good accuracy (86.88%) using 32 PCs as the feature vector and MARS as the classification model.

Denoising of iodine-131 images using a median filter.

OBJECTIVE: An iodine-131 (I) image visually appears to be contaminated with impulse noise. The two-dimensional median filter removes noise without sacrificing the edge information. Its performance depends on the shape and size of the mask. In this study, we have compared the performance of a plus-shape and a square-shape median filter for I whole-body images and found the filter with optimum parameter that improves I image quality acceptable to nuclear medicine physicians. MATERIALS AND METHODS: A total of 150 whole-body I images were exported in DICOM format. These images were converted into PNG format and processed with a plus-shape and a square-shape median filter, with each shape mask having sizes of 3, 5, 7, and 9 pixels. The quality of the processed images was assessed by visual assessment by two nuclear medicine physicians and also quantitatively by evaluating metrics: mutual information, mean square error, peak signal-to-noise ratio, and difference entropy. Nuclear medicine physicians assigned a score to each image on the scale 1 (lowest) to 5 (highest) for image quality on the basis of the noise removal, smoothness, and edge information available in the image. Student's t-test was carried out to find the significant difference in the image quality (α=0.05) between the processed images with square-shape and cross-shape mask with the same pixel size. All experiments including statistical analysis were conducted using R installed on a personal computer. RESULTS: Both median filters improved the image quality of I images. The plus-shape median filter was found to show better performance in comparison with the square-shape median filter (P<0.001). The plus-shape median filter with a mask size of 7 pixels was found to be optimum for the processing of whole-body I images. CONCLUSION: The plus-shape median filter with a mask size of 7 pixels can be used to process whole-body I scintigraphic images without loss of clinical information.

Standardization of image reconstruction parameters for dynamic fluorine-18-fluorodeoxyglucose positron emission tomography/computed tomography.

AIM: The present study aimed to standardize the ordered subset expectation maximization (OSEM) reconstruction parameters for a dynamic PET/CT study. PARTICIPANTS AND METHODS: A locally fabricated phantom was filled with fluorine-18-fluorodeoxyglucose (F-FDG) for four different sphere to background ratios (SBRs), that is, 10 : 1, 8 : 1, 6 : 1, and 4 : 1, and dynamic PET/CT was acquired for 5 min. Transaxial slices were reconstructed using OSEM [full-width at half-maximum (FWHM): 1-7 mm and iterations: 1-8]. Two nuclear medicine physicians visually rated image quality on the basis of the following criteria: score 1: poor quality, score 2: average quality, and score 3: good quality. The quantitative assessment of image quality was performed on the basis of the calculation of noise, horizontal, and vertical line profiles. The standardized parameters were applied to the PET/CT study of seven non-small-cell lung cancer patients, and their image quality was compared with the vendor-provided default parameters. RESULTS: In the phantom study, for SBR 10 : 1, the images reconstructed with FWHM 4 mm and four iterations, for SBR 8 : 1 and 6 : 1, the image with FWHM 3 mm and five iterations, and for SBR 4 : 1, the image with FWHM 2 mm and five iterations were found to have the best quality. In the patient study, FWHM 4 mm and four iterations were found to be suitable for the reconstruction of dynamic F-FDG PET/CT studies with a tumor to background ratio of 10 : 1. With an increase in iterations, noise and sharpness in the image increased, whereas with an increase in FWHM, the image became smoother. CONCLUSION: The standardized reconstruction parameters of OSEM for the dynamic PET/CT study were found to be 4-mm filter FWHM and four iterations in SBR 10 : 1.

Does surgery help in reducing stigma associated with drug refractory epilepsy in children?

INTRODUCTION: Epilepsy has several comorbidities and associated stigma. Stigma associated with epilepsy is well known and prevalent worldwide. Surgical treatment is an established treatment for drug refractory epilepsy. Following surgery in children, it is possible that the stigma may reduce, but such an effect has not been studied earlier. MATERIALS AND METHODS: Analysis of prospectively collected data was performed for pediatric patients at a single tertiary center for treating epilepsy. Child stigma scale, as described by Austin et al., was used to evaluate stigma both pre- and postoperatively. Analysis was done using Paired t test. RESULTS: In this study, following surgery, there was significant reduction of stigma (P<0.001). This was proportional to the reduction in seizures, though there were 9 (30%) patients, who due to persistent neurodisability did not have any reduction of stigma despite having good seizure outcome. CONCLUSION: Surgery in drug-resistant epilepsy helps in reducing stigma. Seizure reduction is probably not the only factor responsible for a change in stigma outcome.

Surgery for Drug-Resistant Epilepsy in Children.

BACKGROUND: Neurosurgical treatment may improve seizures in children and adolescents with drug-resistant epilepsy, but additional data are needed from randomized trials. METHODS: In this single-center trial, we randomly assigned 116 patients who were 18 years of age or younger with drug-resistant epilepsy to undergo brain surgery appropriate to the underlying cause of epilepsy along with appropriate medical therapy (surgery group, 57 patients) or to receive medical therapy alone (medical-therapy group, 59 patients). The patients in the medical-therapy group were assigned to a waiting list for surgery. The primary outcome was freedom from seizures at 12 months. Secondary outcomes were the score on the Hague Seizure Severity scale, the Binet-Kamat intelligence quotient, the social quotient on the Vineland Social Maturity Scale, and scores on the Child Behavior Checklist and the Pediatric Quality of Life Inventory. RESULTS: At 12 months, freedom from seizures occurred in 44 patients (77%) in the surgery group and in 4 (7%) in the medical-therapy group (P<0.001). Between-group differences in the change from baseline to 12 months significantly favored surgery with respect to the score on the Hague Seizure Severity scale (difference, 19.4; 95% confidence interval [CI], 15.8 to 23.1; P<0.001), on the Child Behavior Checklist (difference, 13.1; 95% CI, 10.7 to 15.6; P<0.001), on the Pediatric Quality of Life Inventory (difference, 21.9; 95% CI, 16.4 to 27.6; P<0.001), and on the Vineland Social Maturity Scale (difference, 4.7; 95% CI, 0.4 to 9.1; P=0.03), but not on the Binet-Kamat intelligence quotient (difference, 2.5; 95% CI, -0.1 to 5.1; P=0.06). Serious adverse events occurred in 19 patients (33%) in the surgery group, including hemiparesis in 15 (26%). CONCLUSIONS: In this single-center trial, children and adolescents with drug-resistant epilepsy who had undergone epilepsy surgery had a significantly higher rate of freedom from seizures and better scores with respect to behavior and quality of life than did those who continued medical therapy alone at 12 months. Surgery resulted in anticipated neurologic deficits related to the region of brain resection. (Funded by the Indian Council of Medical Research and others; Clinical Trial Registry-India number, CTRI/2010/091/000525 .).

Practical experience and challenges in the operation of medical cyclotron.

OBJECTIVE: The aim of this article was to share 10 years of operational experience of medical cyclotron and to provide working knowledge on the same. This experience has helped us gain working knowledge on cyclotron operation with practical points, which may help in improving F yield, minimizing the breakdown time, and help in the prevention of the occurrence of unusual events. MATERIALS AND METHODS: Our facility has a self-shielded radioisotope delivery system eclipse 111 medical cyclotron with an 11 MeV proton beam in use for the past 10 years to produce positron emitters - namely, F, N, and F-2 gas - for PET imaging. During F production, we have followed a set protocol comprising the following: monitoring target pressure, rinsing the target with O water just immediately after bombardment, ion source feedback, radiofrequency (RF) feedback, and recording any unusual events that occurred during the operation. Besides this, enrichment of O water, target volume, target current, energy of the beam, variation in argon pressure on the target, bombardment duration, target status (new or old target or total number of previous bombardments on the same target), status of the delivery lines from target to the radiochemistry module (old or new) were also recorded. RESULTS: Rinsing with O water immediately after bombardment increases the life of the target and delivery line. The frequent problems encountered were with the ion source, RF, and target foil rupture. These problems were solved by rebuilding the ion source, changing the fuse of RF, and rebuilding the target. CONCLUSION: F yield can be increased by rinsing with O water immediately after bombardment. The effect of target leak - that is, rupture of vacuum window - can be avoided by immediate stoppage of bombardment.

Preliminary Experience with Yttrium-90-labelled Rituximab (Chimeric Anti CD-20 Antibody) in Patients with Relapsed and Refractory B Cell Non-Hodgkins Lymphoma.

BACKGROUND AND OBJECTIVES: The aim of the study is to evaluate the therapeutic efficacy and safety of Yttrium- 90 radiolabelled chimeric anti CD20 antibody-Rituximab in the treatment of patients with relapsed/ refractory B cell Non-Hodgkins Lymphoma (NHL). METHODS: Twenty patients with relapsed/refractory CD20+ NHL in progressive state were included in the study. These patients had undergone a median of 2 (range 2-5) prior standard chemotherapy ± immunotherapy regimens. All the patients received rituximab 250 mg/m2 on days 1 and 8, and either 14 MBq/kg (0.4 mCi/kg) or 11 MBq/kg (0.3 mCi/kg) of Y-90 Rituximab on day 8 (maximum dose, 32 mCi) depending upon their platelet count. The patients were observed for systemic toxicity and response for at least 12 weeks after therapy. RESULTS: No acute adverse effects were observed after the administration of 90Y-Rituximab. Overall response rate (ORR) was 45% of which complete response (CR) was observed in 2 patients, stable disease in 1 patient and partial response in 6 patients. The therapy was well tolerated with grade IV thrombocytopenia, neutropenia and anemia observed in 3, 4 and 2 patients respectively. CONCLUSION: 90Y-Rituximab therapy is safe and well tolerated in high risk extensively pretreated NHL patients. Toxicity is primarily hematologic, transient and reversible.

Development of a conversion program to make SIMIND-generated SPECT data interfile acceptable to Xeleris.

The Monte Carlo code SIMIND is used in nuclear medicine for research purposes, and also for testing the validity of various applications. Conversion of a SIMIND-generated interfile (header and image data file) is required to process the simulated image data on a Xeleris workstation. Currently there is no conversion program provided with SIMIND to convert its interfile, which is acceptable in any nuclear medicine workstation. Manual editing of the header file is possible with any standard text editor but it is time-consuming (requiring ∼1 h) and stressful. To alleviate this, we have developed a conversion program using FreeMat V4.0 (an open source software similar to MATLAB from MathWorks). It takes an average of 0.04404 s to convert a SIMIND SPECT data interfile and make it acceptable to Xeleris.

Posterior urethral valves: Persistent renin angiotensin system activation after valve ablation and role of pre-emptive therapy with angiotensin converting enzyme-inhibitors on renal recovery.

AIM: To study renin angiotensin system (RAS) activity after posterior urethral valve ablation and the role of early induction of angiotensin converting enzyme-inhibitors (ACE-I) on the outcome of renal function. MATERIALS AND METHODS: Thirty four children underwent valve ablation in which therapy with ACE-I was started 40.5 ± 4.1 (range 32-47 months) formed the study group. Post-ACE-I data were collected after mean duration of 18.2 ± 4.0 (12-28 months). Plasma renin activity (PRA), urinary micro albumin, glomerular filtration rate (GFR), and serum creatinine, before and after therapy were monitored. RESULTS: Therapy with ACE-I resulted in a fall in micro albuminuria by 45.7% and 42.0% in patients without and with vesico ureteral reflux, respectively, and improvement in split renal function by 6.6% and 5.9% GFR respectively. A similar response was noted in patients without and with renal scars. CONCLUSION: The decline in renal function after valve ablation is accompanied by activation of RAS reflected in a gradual rise in PRA. Therapy with ACE-I stabilizes and then improves renal function, thereby, retarding the pace of renal damage.

Is medial temporal injury responsible for pediatric postconcussion syndrome? A prospective controlled study with single-photon emission computerized tomography.

OBJECT: The authors sought to correlate the finding of medial temporal hypoperfusion (MTH) demonstrated on single photon emission computerized tomography (SPECT) with pediatric persistent postconcussion syndrome (PPCS) and to ascertain its usefulness in routine clinical practice. METHODS: Thirty consecutive children with minor head injury and features of PPCS underwent SPECT scanning within 72 hours of injury. Those children having MTH on SPECT were included in the test group (14 patients), and the remaining 16 children comprised the control group. At the end of a 3-month period SPECT scanning was repeated and the incidence of PPCS was assessed clinically in both groups. Repeated SPECT scanning at 3 months revealed persisting MTH in 13 children (93%) in the test group; no child developed MTH in the control group. Twelve children were found to have PPCS in the MTH group compared with only two in the control group, and this was highly statistically significant (relative risk 6.86 [95% confidence interval 1.84-25.51], p = 0.0003). CONCLUSIONS: There exists significant MTH in pediatric patients with PPCS, which would imply that medial temporal lobe damage (involving the hippocampus and related structures) may occur after minor head injury and could be responsible for the symptoms of PPCS observed in this group of patients. Brain SPECT scans may thus help in the early identification of children prone to develop PPCS, and serial SPECT scanning may serve as a platform for testing the efficacy of various neurobehavioral and pharmacological interventions in these patients.