Repeatability and reproducibility study of radiomic features on a phantom and human cohort.

The repeatability and reproducibility of radiomic features extracted from CT scans need to be investigated to evaluate the temporal stability of imaging features with respect to a controlled scenario (test-retest), as well as their dependence on acquisition parameters such as slice thickness, or tube current. Only robust and stable features should be used in prognostication/prediction models to improve generalizability across multiple institutions. In this study, we investigated the repeatability and reproducibility of radiomic features with respect to three different scanners, variable slice thickness, tube current, and use of intravenous (IV) contrast medium, combining phantom studies and human subjects with non-small cell lung cancer. In all, half of the radiomic features showed good repeatability (ICC > 0.9) independent of scanner model. Within acquisition protocols, changes in slice thickness was associated with poorer reproducibility compared to the use of IV contrast. Broad feature classes exhibit different behaviors, with only few features appearing to be the most stable. 108 features presented both good repeatability and reproducibility in all the experiments, most of them being wavelet and Laplacian of Gaussian features.

A prospective study of telephonic contact and subsequent physical follow-up of radically treated lung cancer patients.

BACKGROUND: We tested the hypothesis that telephonic follow-up (FU) may offer a convenient and equivalent alternative to physical FU of radically treated lung cancer patients. DESIGN: Prospective study carried out at a tertiary referral cancer care institute, Mumbai. MATERIALS AND METHODS: Two hundred consecutive lung cancer patients treated with curative intent were followed up regularly with telephonic interviews paired with their routine physical FU visits. Patient satisfaction with the telephonic call and the physical visit, the anxiety level of the patient after meeting the physician and the economic burden of the visit to the patient were noted in a descriptive manner. Kappa statistics was used to assess concurrence between the telephonic and physical impression of disease status. RESULTS: With a median FU duration of 21.5 months, the median satisfaction scores for telephonic and physical FU were 8 and 9, respectively. The prevalence and bias adjusted kappa (PABAK) score of the entire cohort of patients was 0.64 (95% confidence interval [CI] =0.58-0.70). Data analyzed up to first disease progression/relapse on FU had a PABAK score of 0.71 (95% CI = 0.64-0.77) indicating substantial agreement. Patients with disease controlled at the FU had a significant PABAK score of 0.88 (95% CI = 0.80-0.94) indicating excellent concurrence. On average, each patient spent Rs. 5117.10 on travel and Rs. 3079.06 on lodging per FU visit. CONCLUSION: Telephonic FU is substantially accurate in assessing disease status until the first relapse. In a resource-constrained country like India, it is worthwhile to further explore the benefits of such an alternative strategy.

Solitary pulmonary nodule evaluation in regions endemic for infectious diseases: Do regional variations impact the effectiveness of fluorodeoxyglucose positron emission tomography/computed tomography.

BACKGROUND: Fluorodeoxyglucose positron emission tomography/computed tomography (FDG-PET/CT) has become a preferred imaging modality for the evaluation of solitary pulmonary nodule (SPN), particularly in the developed world. Since FDG can concentrate in infective/inflammatory lesions, the diagnostic utility of FDG-PET can be questioned, particularly in regions endemic for infectious decisions. AIM: To evaluate the accuracy of FDG-PET/CT in evaluation of SPNs in a population endemic for infectious disease and to assess if regional variations have an impact on its effectiveness. MATERIALS AND METHODS: All patients who underwent an FDG/PET-CT with a clinico-radiological diagnosis of SPN categorized as indeterminate were included. Based on a maximum standardized uptake values (SUVmax) cut-off of 2.5, lesions were classified as benign (<2.5) or malignant (>2.5) and compared with gold standard histopathology. The diagnostic accuracy of PET-CT to detect malignancy was calculated. On the basis of final histopathology, lesions were grouped as (a) malignant nodules (b) infective/granulomatous nodules with a specific diagnosis and (c) nonspecific inflammatory nodules. The SUVmaxbetween these groups was compared using nonparametric statistical tests. RESULTS: A total of 191 patients (129 males, 62 females) with a median age of 64 years (range: 36-83) were included. Totally, 144 nodules (75.3%) were malignant and 47 were benign (24.7%). Adenocarcinoma (n = 84) was the most common malignancy. Tuberculosis (n = 16) and nonspecific infections (n = 24) were the two most common benign pathologies. There was a significant overlap in the metabolic uptake of malignant (median SUVmax-11.2, range: 3.3-34.6) and tuberculous nodules (median SUVmax-10.3, range: 2.7-22.5) with no statistically difference between their SUVmaxvalues (P = 0.43). The false-positive rate was 65.2% and the false-negative rate was 5.5%. The sensitivity, specificity, positive predictive value, negative predictive value, and accuracy of FDG-PET/CT for detecting malignancy were 94.4%, 34.7%, 81.9%, 66.6%, and 79.5%, respectively. CONCLUSIONS: Though FDG-PET scans show a very high sensitivity for malignant nodules, it has a high false-positive rate and reduced specificity when characterizing SPNs in an infectious endemic region. Physicians must be aware of this limitation in the workup of lung nodules, and regional variations must be considered before further management decisions are taken.

Positron emission tomography thyroid incidentaloma: Is it different in Indian subcontinent?

BACKGROUND: Positron emission tomography (PET) forms an integral part in work-up and follow-up of various malignancies. With the increased use of PET in oncology, finding of an incidental focal thyroid uptake (incidentaloma) is not unusual and presents a diagnostic challenge. AIM: The aim of the following study is to evaluate the frequency and radio-pathologic correlation of focal 18-fluoro deoxyglucose uptake (FDG) on PET within the thyroid from a large series. MATERIALS AND METHODS: Retrospective review of 37,000 consecutive patients who underwent FDG-PET at tertiary cancer center in India. Radiological, pathological, PET scan and follow-up details were evaluated. Statistical analyzes were carried out using Mann Whitney test and Pearson correlation. RESULTS: Abnormal thyroid uptake was seen in 78 (0.2%) patients. Nearly 61 (0.16%) scans had focal and 17 (0.04%) had diffuse FDG uptake. A total of 57 patients with focal uptake were available for further evaluation. No further evaluation was done in 24 (42.1%) patients who had advanced index malignancy. Of the remaining 33 patients 26 were benign and seven were a cause for concern (four primary thyroid cancers, one follicular neoplasm with hurthle cell change and two metastatic cancers). There was no significant correlation in Standardized uptake value (SUV) max of benign and malignant lesion (P = 0.5 on Mann Whitney) or size (r = 0.087 Pearson correlation co-efficient P= 0.667). CONCLUSION: Incidence of PET incidentaloma is low in this large cohort of Indian patients. Nearly 27% of focal incidentaloma were malignant. There was no correlation between the SUVmax, size and malignancy.

Thymic epithelial tumors: Can fluorodeoxyglucose positron emission tomography help in predicting histologic type and stage?

OBJECTIVES: To study the utility of fluorodeoxyglucose (FDG) positron emission tomography (PET) in predicting (1) the World Health Organization (WHO) histologic type and differentiating low-risk from high-risk types. (2) Tumor stage and differentiate early from advanced stage disease. MATERIALS AND METHODS: Patients with thymic epithelial neoplasia who underwent a pretreatment FDG-PET study were included. Tumor maximum standardized uptake value (SUVmax) was correlated with the WHO histologic type and also with the Masaoka-Koga (MK) staging system. Patients with WHO Type A, AB, and B1 were classified as low risk and those with B2 and B3 as high risk. Thymic carcinomas belonged to Type C. Patients with MK Stage I and II disease were grouped as early stage and those with Stage III and IV as an advanced stage. Differences in SUVmax between the various groups were calculated. RESULTS: The SUVmax of thymic carcinomas was significantly higher as compared to low-risk (P = 0.001) and high-risk groups (P = 0.007). The SUVmax of high-risk group was also significantly higher than the low-risk group (P = 0.002). SUVmax cutoff of 6.5 was able to differentiate thymic carcinomas from thymomas with 100% sensitivity and 87.2% specificity. The SUVmax in patients with advanced stage disease showed a higher trend compared to those with early stage, but the difference was not significant (P = 0.167). CONCLUSION: PET can differentiate thymic carcinomas from rest of the thymoma subtypes by the virtue of their higher FDG uptake. It can also provide valuable information in differentiating high-risk from low-risk thymomas and in predicting disease stage.

Utility of PET in unknown primary with cervical metastasis: a retrospective study.

BACKGROUND: FDG-PET is recommended as an investigation in unknown primary tumors, but its definitive role and cost effectiveness are yet to be established. AIMS: dditional value of FDG-PET over conventional imaging in unknown primary tumors with cervical metastasis. SETTING AND DESIGN: Retrospective study in a tertiary level oncology centre. MATERIALS AND METHODS: A total of 112 patients were divided into three groups; 53 with conventional modalities (either computed tomography or magnetic resonance imaging) (group I), 59 with FDG-PET (group II), and group III (subgroup of group II) with both (40 patients). STATISTICAL ANALYSIS: Sensitivity and specificity of both conventional modality and PET were calculated. Association between neck nodes and distant metastasis was analysed using multivariate logistic regression analysis. RESULTS: Sensitivity and specificity for conventional modalities was 92.3% and 50% and sensitivity and specificity of FDG-PET was 92.8% and 71.4%, respectively. FDG-PET detected metastasis in 52.54% of patients. Multivariate logistic regression analysis showed statistically significant association between distant metastasis and multiplicity of nodes (N2b, N2c) (P = 0.007). Among all patients with low neck nodes in group II, FDG-PET detected primaries in 12 patients, 9 of which were infraclavicular (75%). FDG-PET added information to conventional imaging in 32.5% of patients and influenced an overall change in management in 38.9% of patients. CONCLUSIONS: FDG-PET is a valuable tool influencing change of management in unknown primary with cervical metastasis. It is recommended especially in the presence of low or multiple neck nodes in view of high incidence of infraclavicular primary and distant metastasis, respectively.

Isolated left innominate vein aneurysm: a rare cause of mediastinal widening.

Mediastinal widening in an adult is mostly nodal in origin. Occasionally vascular aneurysms may be the underlying cause, in which case the aorta or its branches are most frequently involved. Thoracic venous aneurysms, on the other hand, have been reported only in anecdotes, with fusiform aneurysm of the superior vena cava being the commonest. Isolated aneurysms involving the brachiocephalic/innominate vein are extremely rare. We describe detection of a saccular aneurysm of the innominate vein, as the underlying cause of mediastinal widening seen on a chest radiograph in a 42-year-old asymptomatic woman. The characteristic radiological findings of thoracic venous aneurysms are described with particular reference to the importance of multiplanar computed tomography in such settings. Also discussed is the role of imaging in the diagnosis and guiding the management of this rare entity.

PET-CT in oncology.

Diagnostic imaging plays a pivotal role in cancer management. Imaging has made rapid strides in the past few years with new modalities and new indications getting added all the time. Molecular and metabolic imaging with 18F Flurodeoxyglucose Positron Emission Tomography (FDG PET) and other non-FDG isotopes have revolutionized oncologic imaging and changed our approach towards cancer diagnosis, staging and imaging surveillance. The evidence for the use of FDG PET is mounting every day with newer indications getting added to the literature. The main focus of our discussion will be to understand the basic principles of FDG PET imaging and to study a few of its clinical applications in oncology.

Positron emission tomography / computerized tomography evaluation of primary Hodgkin's disease of liver.

Occurrence of primary Hodgkin's lymphoma (PHL) of the liver is extremely rare. We report on a case of a 60-year-old male who presented with liver mass and B-symptomatology. Hepatoma or hepatic metastasis from a gastrointestinal primary was initially suspected. Tumor markers like AFP, CEA, Total PSA, and CA-19.9 were within normal limits. Positron Emission Tomography / Computerized Tomography (PET/CT) revealed a large hepatic lesion and a nodal mass in the porta hepatis. A liver biopsy was consistent with Hodgkin's lymphoma. There was complete regression of the hepatic lesion and evidence of shrinkage of the nodal mass following four cycles of chemotherapy. 18F Fluro -de-oxy Glucose (FDG) PET / CT in this case helped in establishing a primary hepatic lymphoma by demonstrating the absence of pathologically hypermetabolic foci in any other nodes or organs. PET / CT scan is a useful adjunct to conventional imaging and histopathology, not only to establish the initial diagnosis, but also to monitor treatment response in PHL.

Choroidal metastasis from non-seminomatous germ cell tumour of the testis.

Ureal metastases (although an uncommon site of metastatic involvement) are seen with primary tumours of the lung and breast and, very rarely, in other primary tumours. Undescended testis predisposes to malignancy. We present a rare case of choroidal metastasis from a non-seminomatous germ cell tumour arising in an undescended testis, and describe its clinical, ultrasound and MRI findings together with a relevant review of the literature.

Diffuse-type giant cell tumor of the subcutaneous thigh.

Diffuse-type giant cell tumor is an extra-articular form of pigmented villonodular synovitis. The localized form of this lesion (tenosynovial giant cell tumor) is frequent, representing the most common subset arising from the synovium of a joint, bursa or tendon sheath, with 85% of cases occurring in the fingers. The less frequent diffuse-type giant cell tumors are commonly located in the periarticular soft tissues, but on rare occasions these lesions can be purely intramuscular or subcutaneous We report the case of a 26-year-old female with diffuse-type giant cell tumor of the subcutaneous thigh, remote from a joint, bursa or tendon sheath. A review of the literature did not reveal any similar description of a diffuse-type giant cell tumor completely within the subcutaneous thigh, remote from a joint, bursa or tendon sheath. These lesions were initially regarded as inflammatory or reactive processes, but since the identification of clonal abnormalities in these patients, and in view of their capacity for autonomous growth, they are now widely considered to represent benign neoplasms.

Bilateral persistent hyperplastic primary vitreous.

We report a case of bilateral persistent hyperplastic primary vitreous (PHPV) in a 5-month-old infant who presented with bilateral leukokoria. The child was referred for ocular ultrasound with a clinical suspicion of retinoblastoma. Grey-scale evaluation revealed an echogenic band in the posterior segment of the left globe extending from the posterior surface of the lens capsule to the optic disc. Doppler examination revealed the presence of blood flow in the band. Ultrasound assessment of the contralateral globe showed an elevated mass of echogenic tissue in the posterior segment, in contact with the optic disc. Most cases of PHPV are sporadic and unilateral. Bilateral PHPV is rare. In a study by Pollard of 83 cases, only two patients (2.4%) had bilateral PHPV. The imaging features in this case point toward the diagnosis of bilateral PHPV. We suggest that this entity, although rare, should be considered in the differential diagnosis while evaluating bilateral leukokoria.

Primary rhabdomyosarcoma of the seminal vesicle.

Sarcomas of the seminal vesicle are very rare and poorly documented; as it is not always possible to pinpoint a truly vesicular origin of the pelvic mass due to local spread at the time of presentation. The purpose of the article is to document and characterize a rhabdomyosarcoma of the seminal vesicle of which to the knowledge of the authors there has been no previous report in the English literature.