[Positron emission tomography (PET) in oncology (Part I)]

A serial image analysis architecture with positron emission tomography using machine learning combined for the detection of lung cancer

INTRODUCTION AND OBJECTIVES

Lung cancer is the second type of cancer with the second highest incidence rate and the first with the highest mortality rate in the world. Machine learning through the analysis of imaging tests such as positron emission tomography/computed tomography (PET/CT) has become a fundamental tool for the early and accurate detection of cancer. The objective of this study was to propose an image analysis architecture (PET/CT) ordered in phases through the application of ensemble or combined machine learning methods for the early detection of lung cancer by analyzing PET/CT images.

MATERIAL AND METHODS

A retrospective observational study was conducted utilizing a public dataset entitled "A large-scale CT and PET/CT dataset for lung cancer diagnosis." Various imaging modalities, including CT, PET, and fused PET/CT images, were employed. The architecture or framework of this study comprised the following phases: 1. Image loading or collection, 2. Image selection, 3. Image transformation, and 4. Balancing the frequency distribution of image classes. Predictive models for lung cancer detection using PET/CT images included: a) the Stacking model, which used Random Forest and Support Vector Machine (SVM) as base models and complemented them with a logistic regression model, and b) the Boosting model, which employed the Adaptive Boosting (AdaBoost) model for comparison with the Stacking model. Quality metrics used for evaluation included accuracy, precision, recall, and F1-score.

RESULTS

This study showed a general performance of 94% with the Stacking method and a general performance of 77% with the Boosting method.

CONCLUSIONS

The Stacking method proved to be a model with high performance and quality for lung cancer detection when analyzing PET/CT images.

18F-FDG PET/CT in ovarian cancer recurrence: Clinical impact, correlation with ceCT and CA-125, and prognostic value

AIM

To evaluate ¹⁸F-FDG-PET/CT for suspected ovarian cancer relapse with negative/inconclusive conventional imaging, or restaging potentially resectable ovarian cancer relapse.

MATERIAL AND METHODS

Thirty-six cases and 140 locations were studied. PET/CT, ceCT and serum CA-125 was conducted in all cases. Nineteen cases were requested for restaging, 17 for suspected relapse. We compared ceCT and PET/CT, assessed by histopathology or radiological follow-up, calculating sensitivity (S) and positive predictive value (PPV) by cases and lesions. We evaluated the correlation between size, number, uptake of the lesions and CA-125. We conducted survival analysis, using ROC curves to calculate the optimal cut-off of SUVmax for survival prediction. We checked whether PET/CT modify the therapeutic attitude vs. conventional imaging.

RESULTS

PET/CT and ceCT were concordant in 12 cases: 11 positives (30 lesions), all confirmed. There was 1 FN. In the 24 non-concordant, PET/CT was positive in 19 (97 lesions); ceCT in 21 (59 lesions); 54% of the lesions were concordant. Overall, PET/CT detected 127 lesions, with S = 97% and PPV = 100%. ceCT detected 89 lesions, with S = 61% and PPV = 90%. No significant correlation was found between CA-125 and the other parameters. PET/CT detected 10 positive cases, with normal CA-125. PET/CT modified therapeutic management in 15 cases. Significant differences were found in survival with SUVmax = 11.8 CONCLUSIONS: PET/CT plays an important role in ovarian cancer relapse, with sensitivity and PPV higher than ceCT, modified therapeutic management in up to 42% of cases, and could be a valuable tool for predicting survival.

18F-FDG PET/CT in ovarian cancer recurrence: Clinical impact, correlation with ceCT and CA-125, and prognostic value

AIM

To evaluate ¹⁸F-FDG-PET/CT for suspected ovarian cancer relapse with negative/inconclusive conventional imaging, or restaging potentially resectable ovarian cancer relapse.

MATERIAL AND METHODS

Thirty-six cases and 140 locations were studied. PET/CT, ceCT and serum CA-125 was conducted in all cases. Nineteen cases were requested for restaging, 17 for suspected relapse. We compared ceCT and PET/CT, assessed by histopathology or radiological follow-up, calculating sensitivity (S) and positive predictive value (PPV) by cases and lesions. We evaluated the correlation between size, number, uptake of the lesions and CA-125. We conducted survival analysis, using ROC curves to calculate the optimal cut-off of SUVmax for survival prediction. We checked whether PET/CT modify the therapeutic attitude vs. conventional imaging.

RESULTS

PET/CT and ceCT were concordant in 12 cases: 11 positives (30 lesions), all confirmed. There was 1 FN. In the 24 non-concordant, PET/CT was positive in 19 (97 lesions); ceCT in 21 (59 lesions); 54% of the lesions were concordant. Overall, PET/CT detected 127 lesions, with S=97% and PPV=100%. ceCT detected 89 lesions, with S=61% and PPV=90%. No significant correlation was found between CA-125 and the other parameters. PET/CT detected 10 positive cases, with normal CA-125. PET/CT modified therapeutic management in 15 cases. Significant differences were found in survival with SUVmax=11.8 CONCLUSIONS: PET/CT plays an important role in ovarian cancer relapse, with sensitivity and PPV higher than ceCT, modified therapeutic management in up to 42% of cases, and could be a valuable tool for predicting survival.

[Validation of segmentation techniques for positron emission tomography using ex-vivo images of oncological surgical specimens]

OBJECTIVE

To design a novel ex-vivo acquisition technique to establish a common framework to validate different segmentation techniques for oncological PET images. To evaluate several automatic segmentation algorithms on this set of images.

MATERIAL AND METHODS

In 15 patients with cancer, ex-vivo PET studies of surgical specimens removed during surgery were performed after injection of (18)F-FDG. Images were acquired in two scanners: a clinical PET/CT and a high-resolution PET scanner. Real tumor volume was determined in each patient, and a reference image was generated for segmentation of each tumor. Images were segmented with 12 automatic algorithms and with a standard method for PET (relative threshold at 42%) and results were evaluated by quantitative parameters.

RESULTS

It has been possible to demonstrate by segmentation of PET images of surgical specimens that on high resolution PET images, 8 out of 12 evaluated segmentation techniques outperformed the standard method, whose value is 42%. However, none of the algorithms outperformed the standard method when applied on images from the clinical PET/CT. Due to the great interest of this set of PET images, all studies have been published on the Internet in order to provide a common framework for validation and comparison of different segmentation techniques.

CONCLUSIONS

We have proposed a novel technique to validate segmentation techniques for oncological PET images, acquiring ex-vivo PET studies of surgical specimens. We have demonstrated the usefulness of this set of PET images by evaluating several automatic segmentation algorithms.

[Findings of the (18)F-FDG PET-CT in a cardiac angiosarcoma complicated by a cardiac rupture]

Primary malignant tumors of the heart are a rare condition. The most common type is the cardiac angiosarcoma. The symptoms of this disease are very nonspecific and can be very difficult to diagnose by conventional imaging techniques. We report the case of a male patient with cardiac angiosarcoma who also had a rare complication, this being cardiac rupture, which required the use of (18)F-FDG PET-CT to demonstrate the mass malignancy and to reach a definitive diagnosis.

[Positron emission tomography-computed tomography in tumors of the locomotor apparatus]

Positron emission tomography/computed tomography (PET/CT) is a hybrid imaging technique that combines the anatomic information from CT with the metabolic information acquired from PET after the administration of specific radiotracers, the most commonly used of which is F18-fluorodeoxyglucose (FDG). In oncology, this technique is based on the increased uptake of FDG by malignant lesions. In the locomotor apparatus, some uptake by bones and soft tissues is physiological or benign and this uptake must be differentiated from uptake by malignancies, whether primary or secondary. The most important limitations are active inflammatory or infectious processes, which are positive on PET images, and malignant lesions that are smaller than 1cm, cystic, necrotic, or low-grade, which are negative on PET images. PET/CT in the locomotor apparatus is especially useful for the detection of metastases from the most common tumors. It is also used for staging and monitoring the response to treatment of some hematological tumors like lymphoma, where it is fundamental to determine whether the bone marrow has been infiltrated, or myeloma. Lastly, although it is not yet an established indication, PET/CT is being increasingly used to study sarcomas, because it can provide additional information that can be useful for the characterization and grading of tumors, for guiding biopsies, for staging and re-staging, and for evaluating the response to neoadjuvant therapy as well as for evaluating new drugs in clinical trials.

[Variations of the hepatic SUV in relation to the body mass index in whole body PET-CT studies]

OBJECTIVE

To evaluate SUV changes in the liver in relation to body mass index (BMI) of patients who undergo whole body PET-CT scans.

MATERIAL AND METHODS

A retrospective study was performed. The variables studied were injected dose of (18F)FDG (mCi), age (years), blood glucose level (mg/dL), height (cm) and weight (kg). BMI was calculated and the SUV mean value was expressed according to gender and BMI. A linear regression analysis was applied to identify the independent variables that best predict the SUV value.

RESULTS

Six hundred and three patients were studied (305 women, 298 men; mean age: 54.9±15.2 years old). Mean SUV measurement was significantly higher in males than females and increased significantly both in male and female patients who were overweight and even more in obese patients. The independent variables that best predicted the SUV value were gender, age, and BMI. In those patients having similar characteristics related to the analyzed variables, the SUV value increased by 0.002 for each increase in one year, and by 0.066 per unit increase in the BMI value.

CONCLUSIONS

Hepatic uptake of (18F)FDG increases according to the patient's BMI. The independent variables that best predict the hepatic SUV value are age and sex of patients. Our findings show that the practice of using the physiological hepatic metabolic activity level as a reference regarding questionable deposits elsewhere in the abdomen and pelvis is not useful, at least in male patients with overweightness and obesity.

Utilidad de la PET con FDG en el estudio de estadificación inicial, recurrencia y respuesta al tratamiento con imatinib (Glivec) en pacientes diagnosticados de un tumor del estroma gastrointestinal

INTRODUCTION

Gastrointestinal stromal tumors (GISTs) account for almost 4 % of all gastrointestinal neoplasms. Recently, a new type of tyrosine kinase inhibitor (Glivec), has been successfully used in patients with metastasic or unresectable disease. The aim of the study is to show the utility of PET in the staging, recurrence and treatment response to Glivec in GIST tumors.

MATERIALS AND METHODS

48 whole body FDG-PET studies in 27 patients with GIST (19 men/mean age = 56 y) were evaluated for initial staging (n = 13), recurrence (n = 15) or treatment response to Glivec (n = 20). Images were acquired in a whole body 2D mode using attenuation correction on an Advance Nxi G.E.MS camera and were evaluated visually and quantatively using SUV analysis. Results were compared with radiological findings, hystological confirmation or follow-up.

RESULTS

In the initial staging evaluation, FDG-PET shows a more extensive disease than suspected in 3/10 patients. In other 3 patients PET ruled out mesenteric or peritoneal disease. In the evaluation of treatment response to Glivec, FDG-PET showed a good response in eleven patients (complete response in seven and partial response in four). In this group a sixty percent decrease of the SUV max was assessed. Two patients showed no response to Glivec at doses of 400 mg or 800 mg, showing a stable SUV value and/or increased in some abdominal lesions. PET detected recurrence in one patient.

CONCLUSIONS

This study show how FDG-PET is accurate in the early treatment response to Glivec. PET could be helpful in the staging and recurrence of GIST tumors.

[Value of thallium 201-SPECT in typing brain space-occupying lesions]

OBJECTIVE

This study was designed to assess the diagnostic accuracy of single photon emission computed tomography with Tl-201 (SPECT Tl-201) to establish the tumoral or non-tumoral nature of brain space occupying lesions in comparison with usual diagnostic techniques.

MATERIAL AND METHODS

The study population consisted of 37 patients, 24 men (64.9 %) and 13 women (35.1 %), mean age 48 +/- 16 years. After establishing the clinical and radiological diagnosis of brain lesion, all patients underwent SPECT Tl-201, evaluating it only by subjective analysis and blinded to neuroestructural techniques. After surgical resection all patients were evaluated anatomopathologically to establish the histologic nature.

RESULTS

The sensitivity of SPECT Tl-201 (0.87) was higher than standard neuroimaging techniques (0.78). Specificity (0.43), positive (0.87) and negative (0.43) predictive values of SPECT were similar to neuroestructural procedures (MRI and CT scan) with 0.43, 0.82 and 0.38 values. Tumoral disease prevalence was 0.81. Neuroestructural procedures were non-conclusive in 18.9 % of the studies. No non-conclusive results were obtained with SPECT Tl-201.

CONCLUSIONS

SPECT Tl-201 is a diagnostic procedure of high sensitivity to establish the tumoral nature of brain lesions, with poor specificity, similar to structural X-ray techniques.

Detección temprana con PET-FDG de tumores inesperados de colon en pacientes con otra patología tumoral conocida

Patients with known tumors can have other tumors in unexpected places, the so-called synchronic or second primary cancers. In these patients, whole-body fluorodeoxyglucose positron emission tomography (FDG-PET) to study known tumors can identify other tumors that would escape notice if only the sites of known tumors were investigated. We present 3 patients in whom whole-body FDG-PET was requested to study known tumors: 2 lung cancers and one melanoma. FDG-PET localized the known lesions and revealed hypermetabolic uptake in the abdomen in all 3 patients. Endoscopic polypectomy and pathologic analysis were performed, revealing 2 colon cancers and a tubular adenoma with a high degree of focal atypia. Whole body FDG-PET is a highly useful technique in the early diagnosis of colorectal cancer, which in turn allows complete cure of patients with this disease.