Interpretation Of Scans Research Articles

Artificial intelligence-based analysis of whole-body bone scintigraphy: The quest for the optimal deep learning algorithm and comparison with human observer performance

PurposeWhole-body bone scintigraphy (WBS) is one of the most widely used modalities in diagnosing malignant bone diseases during the early stages. However, the procedure is time-consuming and requires vigour and experience. Moreover, interpretation of WBS scans in the early stages of the disorders might be challenging because the patterns often reflect normal appearance that is prone to subjective interpretation. To simplify the gruelling, subjective, and prone-to-error task of interpreting WBS scans, we developed deep learning (DL) models to automate two major analyses, namely (i) classification of scans into normal and abnormal and (ii) discrimination between malignant and non-neoplastic bone diseases, and compared their performance with human observers. Materials and MethodsAfter applying our exclusion criteria on 7188 patients from three different centers, 3772 and 2248 patients were enrolled for the first and second analyses, respectively. Data were split into two parts, including training and testing, while a fraction of training data were considered for validation. Ten different CNN models were applied to single- and dual-view input (posterior and anterior views) modes to find the optimal model for each analysis. In addition, three different methods, including squeeze-and-excitation (SE), spatial pyramid pooling (SPP), and attention-augmented (AA), were used to aggregate the features for dual-view input models. Model performance was reported through area under the receiver operating characteristic (ROC) curve (AUC), accuracy, sensitivity, and specificity and was compared with the DeLong test applied to ROC curves. The test dataset was evaluated by three nuclear medicine physicians (NMPs) with different levels of experience to compare the performance of AI and human observers. ResultsDenseNet121_AA (DensNet121, with dual-view input aggregated by AA) and InceptionResNetV2_SPP achieved the highest performance (AUC = 0.72) for the first and second analyses, respectively. Moreover, on average, in the first analysis, Inception V3 and InceptionResNetV2 CNN models and dual-view input with AA aggregating method had superior performance. In addition, in the second analysis, DenseNet121 and InceptionResNetV2 as CNN methods and dual-view input with AA aggregating method achieved the best results. Conversely, the performance of AI models was significantly higher than human observers for the first analysis, whereas their performance was comparable in the second analysis, although the AI model assessed the scans in a drastically lower time. ConclusionUsing the models designed in this study, a positive step can be taken toward improving and optimizing WBS interpretation. By training DL models with larger and more diverse cohorts, AI could potentially be used to assist physicians in the assessment of WBS images.

MRRead—A Novel Approach to Training Residents in Magnetic Resonance Image Temporomandibular Joint Interpretation

The purpose of our study was to create an online, web-based training module that would instruct a group of participants in the interpretation of a magnetic resonance image (MRI) of the temporomandibular joint (TMJ) scan in a logical, step-wise manner to locate and identify all relevant features of internal derangement. The investigator's hypothesis was that implementing the MRRead TMJ training module would improve the participants' competency in the interpretation of MRI TMJ scans. The investigators designed and implemented a single-group prospective cohort study. The study population was composed of oral and maxillofacial surgery interns, residents, and staff. Subjects eligible for study inclusion were oral and maxillofacial surgeons of any level, between 18 and 50years of age, that completed the MRRead training module to completion. The primary outcome variable was the difference between the pretest and post-test scores of the participants, and the frequency of missing internal derangement findings before and after the course. Secondary outcomes of interest were subjective data gathered from the course, including participant feedback as well as subjective evaluation of the training module and perceived benefit, as well as the learner's self-reported confidence level in interpreting MRI TMJ scans on their own before and after completion of the course. Descriptive and bivariate statistics were used. The study sample consisted of 68 subjects, aged 20 to 47 (M=29.1) years. When comparing the results of the exams precourse and post course, the overall frequency of missed features of internal derangement decreased from 19.7 to 5.9, and the total score overall increased from 8.5 to 68.6%. Regarding secondary outcomes, the majority of participants indicated that they agree or strongly agree with a number of positive subjective questions asked. In addition, there was a statistically significant increase in the participants' comfort levels in the interpretation of MRI TMJ scans. The results of this study confirm the hypothesis that completing the MRRead training module (www.MRRead.ca) improves competency and comfort among participants in the interpretation of MRI TMJ scans and their identification of features of internal derangement correctly.

PET Criteria by Cancer Type from Imaging Interpretation to Treatment Response Assessment: Beyond FDG PET Score.

in recent years, the role of positron emission tomography (PET) and PET/computed tomography (PET/CT) has emerged as a reliable diagnostic tool in a wide variety of pathological conditions. This review aims to collect and review PET criteria developed for interpretation and treatment response assessment in cases of non-[18F]fluorodeoxyglucose ([18F]FDG) imaging in oncology. A wide literature search of the PubMed/MEDLINE, Scopus and Google Scholar databases was made to find relevant published articles about non-[18F]FDG PET response criteria. The comprehensive computer literature search revealed 183 articles. On reviewing the titles and abstracts, 149 articles were excluded because the reported data were not within the field of interest. Finally, 34 articles were selected and retrieved in full-text versions. available criteria are a promising tool for the interpretation of non-FDG PET scans, but also to assess the response to therapy and therefore to predict the prognosis. However, oriented clinical trials are needed to clearly evaluate their impact on patient management.

LUNG CANCER DETECTION USING MORPHOLOGICAL WATERSHED OPERATIONS

Cancer is one of the most serious and widespread disease across the world accounting for a large number of deaths every year. Lung cancer is the leading cause of cancer deaths in the world, having high mortality rate. The main problem in lung cancer is that most of these cases are diagnosed at the later stages of cancer, making treatments more problematic and reducing the survival chances. The survival rate of Cancer patients’ increment from 14-49% if the illness is recognized in time. Early discovery of lung cancer can expand the possibility of survival among individuals. A CT scan is more likely to show lung tumors than routine chest x-rays as it can show the size, shape, and position of any lung tumor and can help find enlarged lymph nodes that might contain cancer that has spread. Visual interpretation of CT scan can be difficult and error prone. Hence image processing techniques are used to detect the cancer nodules at an early stage. This paper deals with an automated approach for early detection of lung cancer. This algorithm is proposed using methods such as Median Filtering for Image Smoothing, Contrast Adjustment for enhancing the image, Segmentation using Morphological Watershed Operations and Otsu’s Thresholding. Key Words: CT, Median Filtering, Contrast Adjustment, Morphological Watershed Operations, Otsu Thresholding.

Outcomes From More Than 1 Million People Screened for Lung Cancer With Low-Dose CT Imaging

Lung cancer screening (LCS) with low-dose CT (LDCT) imaging was recommended in 2013, making approximately 8 million Americans eligible for LCS. The demographic characteristics and outcomes of individuals screened in the United States have not been reported at the population level. What are the outcomes among people screened and entered in the American College of Radiology's Lung Cancer Screening Registry compared with those of trial participants? This was a cohort study of individuals undergoing baseline LDCT imaging for LCS between 2015 and 2019. Predictors of adherence to annual screening were computed. LDCT scan interpretations by Lung Imaging Reporting and Data System (Lung-RADS) score, cancer detection rates (CDRs), and stage at diagnosis were compared with National Lung Cancer Screening Trial data. Adherence was 22.3%, and predictors of poor adherence included current smoking status and Hispanic or Black race. On baseline screening, 83%of patients showed negative results and 17%showed positive screening results. The overall CDR was 0.56%. The percentage of people with cancer detected at baseline was higher in the positive Lung-RADS categories at 0.4%for Lung-RADS category 3, 2.6%for Lung-RADS category 4A, 11.1%for Lung-RADS category 4B, and 19.9%for Lung-RADS category 4X. The cancer stage distribution was similar to that observed in the National Lung Cancer Screening Trial, with 53.5%of patients receiving a diagnosis of stage I cancer and 14.3%with stage IV cancer. Underreporting into the registry may have occurred. This study revealed both the positive aspects of CT scan screening for lung cancer and the challenges that remain. Findings on CT imaging were correlated accurately with lung cancer detection using the Lung-RADS system. A significant stage shift toward early-stage lung cancer was present. Adherence to LCS was poor and likely contributes to the lower than expected cancer detection rate, all of which will impact the outcomes of patients undergoing screening for lung cancer.

High Interobserver Agreement on PSMA PET/CT Even in the Absence of Clinical Data.

Recommended by current guidelines, prostate-specific membrane antigen (PSMA)-directed PET/CT is increasingly used in men with prostate cancer (PC). We aimed to provide concordance rates using the PSMA reporting and data system (RADS) for scan interpretation and also determine whether such agreement rates are affected by available patient characteristics at time of scan. Sixty men with PC, who all underwent 68Ga-PSMA-11 PET/CT, were included. Three independent, experienced readers indicated general scan parameters (including overall scan result, organ or lymph node [LN] involvement, and appropriateness of radioligand therapy). Applying PSMA-RADS 1.0, observers also had to conduct RADS scoring on a target lesion (TL) and overall scan level. During the first read, observers were masked to all relevant clinical information, whereas on a second read, relevant patient characteristics were displayed, thereby allowing for determination of impact of available clinical information for scan interpretation. We used intraclass correlation coefficients (ICCs; with 95% confidence intervals [CIs]), which were then rated according to Cicchetti (0.4-0.59 fair, 0.6-0.74 good, and 0.75-1 excellent agreement). For general parameters, agreement rates were excellent, including an overall scan result (ICC, 0.85; 95% CI, 0.76-0.90), LN metastases (ICC, 0.89; 95% CI, 0.83-0.93), organ involvement (ICC, 0.82; 95% CI, 0.72-0.89), and indication for radioligand therapy (ICC, 0.94; 95% CI, 0.90-0.96). Overall RADS scoring was also excellent with an ICC of 0.91 (95% CI, 0.96-09.4). On a TL-based level, 251 different lesions were selected by the 3 observers (with 73 chosen by all 3 readers). RADS-based concordance rates were fair to excellent: all lesions, ICC of 0.78 (95% CI, 0.67-0.85); LN, ICC of 0.81 (95% CI, 0.63-0.92); skeleton, ICC of 0.55 (95% CI, 0-0.84); and prostate, ICC of 0.48 (95% CI, 0.17-0.78). When performing a second read displaying patient's characteristics, there were only minor modifications to the previously applied RADS scoring on a TL-based level (overall, n = 8): each reader 1 and 2 in 3/60 (5%) instances, and reader 3 in 2/60 (3.3%) instances. The main reason for recategorization (mainly upstaging) was provided information on PSA levels (4/8, 50%). Applying PSMA-RADS, concordance rates were fair to excellent, whereas relevant modifications were rarely observed after providing clinical data. As such, even in the absence of patient information, standardized frameworks still provide guidance for reading PSMA PETs. Those findings may have implications for a high throughput in a busy PET practice, where patient details cannot always be retrieved at time of scan interpretation or in the context of clinical trials or central reviews in which readers may be blinded to clinical data.

Deep learning of renal scans in children with antenatal hydronephrosis

Antenatal hydronephrosis (ANH) is one of the most common anomalies identified on prenatal ultrasound, found in up to 4.5% of all pregnancies. Children with ANH are surveilled with repeated renal ultrasound and when there is high suspicion for a ureteropelvic junction obstruction on renal ultrasound, a mercaptuacetyltriglycerine (MAG3) Lasix renal scan is performed to evaluate for obstruction. However, the challenging interpretation of MAG3 renal scans places patients at risk of misdiagnosis. Our objective was to analyze MAG3 renal scans using machine learning to predict renal complications. We hypothesized that our deep learning model would extract features from MAG3 renal scans that can predict renal complications in children with ANH. We performed a case-control study of MAG3 studies drawn from a population of children with ANH concerning for ureteropelvic junction obstruction evaluated at our institution from January 2009 until June of 2021. The outcome was renal complications that occur ≥6 months after an equivocal MAG-3 renal scan. We created two machine learning models: a deep learning model using the radiotracer concentration versus time data from the kidney of interest and a random forest model created using clinical data. The performance of the models was assessed using measures of diagnostic accuracy. We identified 152 eligible patients with available images of which 62 were cases and 90 were controls. The deep learning model predicted future renal complications with an overall accuracy of 73% (95% confidence inteveral [CI] 68-76%) and an AUC of 0.78 (95% CI 0.7, 0.84). The random forest model had an accuracy of 62% (95% CI 60-66%) and an AUC of 0.67 (95% CI. 0 64, 0.72) DISCUSSION: Our deep learning model predicted patients at high risk of developing renal complications following an equivocal renal scan and discriminate those at low risk with moderately high accuracy (73%). The deep learning model outperformed the clinical model built from clinical features classically used by urologists for surgical decision making. Our models have the potential to influence clinical decision making by providing supplemental analytical data from MAG3 scans that would not otherwise be available to urologists. Future multi-institutional retrospective and prospective trials are needed to validate our model.

Absceso hepático piógeno roto: reporte de caso

Liver abscess is an uncommon pathology caused by the invasion of amebic, pyogenic or fungal microorganisms into the healthy parenchyma of the liver. We present the clinical case of a 27-year-old male adult with alcoholism and occasional smoking, with no other associated comorbidities, who was admitted to the General Hospital of Tecamachalco, Puebla, Mexico, for generalized abdominal pain and intolerance to oral administration for 4 days. CT scan interpretation showed data suggestive of hepatic abscess in segments VI and VII of the liver with abundant fluid in the abdomino-pelvic cavity. Exploratory laparotomy was performed with the finding of rupture of a ruptured pyogenic liver abscess, drainage and cleaning of the abdominal cavity was performed, antimicrobial therapy was started with satisfactory evolution and discharge without complications. In this patient there were no findings of lesion or cause of migration of the infection from another abdominal organ to the liver, confirming the unknown origin.

Validation of Deauville Score for Response Evaluation in Hodgkin's Lymphoma.

Positron emission tomography (PET) using F-18 fluorodeoxyglucose (FDG) for treatment monitoring in patients with lymphoma is one of the most well-developed clinical applications. Deauville five-point score (DS) is recommended for response assessment in international guidelines. DS gives the threshold for adequate or inadequate response to be adapted according to the clinical context or research question. We aimed to validate DS in Hodgkin's lymphoma (HL) by retrospectively assigning this score to F-18 FDG PET-computed tomography (CT) studies done before 2016 and analyzing its concordance with the line of management. The secondary aim was to assess the reproducibility of DS in the interpretation of PET-CT scans. A total of 100 eligible consecutive patients underwent F-18 FDG PET-CT scans between January 2014 and December 2015. Their interim, end of treatment, and follow-up PET scans were retrospectively visually analyzed and assigned DS by three nuclear medicine physicians. Concordance was defined as agreement between the DS assigned and the line of treatment. Interobserver variability was calculated using weighted Kappa and presented with 95% confidence interval. Among 212 scans assigned DS, 165 scans showed agreement between the DS and line of treatment. Of these, 95.2% of scans scored DS 1-3 were kept on following or the same treatment plan was continued and patients did well. Among the scans that showed discordance, 24 scans scored DS 4/5 were continued on the same treatment regimen and the next assessment showed disease progression. Our study confirmed that DS is a useful tool to aid in reporting F-18 FDG PET-CT in the management of HL with good positive and negative predictive values. This study also demonstrated good interobserver agreement.

Detecting Historical Terrain Anomalies With UAV-LiDAR Data Using Spline-Approximation and Support Vector Machines

The documentation of historical remains and cultural heritage is of great importance to preserve historical knowledge. Many studies use low-resolution airplane-based laser scanning and manual interpretation for this purpose. In this study, a concept to automatically detect terrain anomalies in a historical conflict landscape using high-resolution UAV-LiDAR data was developed. We applied different ground filter algorithms and included a spline-based approximation step in order to improve the removal of low vegetation. Due to the absence of comprehensive labeled training data, a one-class support vector machine algorithm was used in an unsupervised manner in order to automatically detect the terrain anomalies. We applied our approach in a study site with different densities of low vegetation. The morphological ground filter was the most suitable when dense near-ground vegetation is present. However, with the use of the spline-based processing step, all filters used could be significantly improved in terms of the F1-score of the classification results. It increased by up to 42 percentage points in the area with dense low vegetation and by up to 14 percentage points in the area with sparse low vegetation. The completeness (recall) reached maximum values of 0.8 and 1.0, respectively, when taking into account the results leading to the highest F1-score for each filter. Therefore, our concept can support on-site field prospection.

Effects of Hydrothermal Treatment on Mesopore Structure and Connectivity in Doped Ceria-Zirconia Mixed Oxides.

Pore size and pore connectivity control diffusion-based transport in mesopores, a crucial property governing the performance of heterogeneous catalysts. In many cases, transition-metal oxide catalyst materials are prepared from molecular precursors involving hydrothermal treatment followed by heat treatment. Here, we investigate the effects of such a hydrothermal aftertreatment step, using an aqueous ammonia solution, on the disordered mesopore network of CexZr1-x-y-zYyLazO2-δ mixed oxides. This procedure is a common synthesis step in the preparation of such ceria-based oxygen storage materials applied in three-way catalysis, employed to improve the materials' thermal stability. We perform state-of-the-art Ar-physisorption analysis, especially advanced hysteresis scanning, to paint a detailed picture of the alterations in mesopore space caused by the hydrothermal aftertreatment and subsequent aging at 1050 °C. Furthermore, we investigate the network characteristics by electron tomography in combination with suitable statistical analysis, enabling a consistent interpretation of the desorption scans (physisorption). The results indicate that the hydrothermal aftertreatment enhances the mesopore connectivity of the continuous 3D network by widening pores and especially necks, hence facilitating accessibility to the particles' internal surface area and the ability to better withstand high temperatures.

Computer-Aided Diagnosis of Pulmonary Nodules in Rheumatoid Arthritis.

(1) Background: Rheumatoid arthritis (RA) is considered a systemic inflammatory pathology characterized by symmetric polyarthritis associated with extra-articular manifestations, such as lung disease. The purpose of the present study is to use CAD in the detection of rheumatoid pulmonary nodules. In addition, we aim to identify the characteristics and associations between clinical, laboratory and imaging data in patients with rheumatoid arthritis and lung nodules. (2) Methods: The study included a number of 42 patients diagnosed with rheumatoid arthritis according to the 2010 American College of Rheumatology (ACR)/European League Against Rheumatism (EULAR) criteria, examined from January 2017 to November 2022 in the Departments of Rheumatology and Radiology and Medical Imaging of the University of Medicine and Pharmacy of Craiova. Medical records were reviewed. A retrospective blinded review of CT for biopsy-proven pulmonary nodules in RA using Veolity LungCAD software was performed (MeVis Medical Solutions AG, Bremen, Germany). Imaging was also reviewed by a senior radiologist. (3) Results: The interobserver agreement proved to be moderate (κ = 0.478) for the overall examined cases. CAD interpretation resulted in false positive results in the case of 12 lung nodules, whereas false negative results were reported in the case of 8 lung nodules. The mean time it took for the detection of lung nodules using CAD was 4.2 min per patient, whereas the detection of lung nodules by the radiologist was 8.1 min per patient. This resulted in a faster interpretation of lung CT scans, almost reducing the detection time by half (p < 0.001). (4) Conclusions: The CAD software is useful in identifying lung nodules, in shortening the interpretation time of the CT examination and also in aiding the radiologist in better assessing all the pulmonary lung nodules. However, the CAD software cannot replace the human eye yet due to the relative high rate of false positive and false negative results.

Maxillofacial Fracture Detection Using Transfer Learning Models : A Review

Early detection and treatment of face bone fractures reduce long-term problems. Fracture identification needs CT scan interpretation, but there aren't enough experts. To address these issues, researchers are classifying and identifying objects. Categorization-based studies can't pinpoint fractures. Proposed Study Convolutional neural networks with transfer learning may detect maxillofacial fractures. CT scans were utilized to retrain and fine-tune a convolutional neural network trained on non-medical images to categorize incoming CTs as "Positive" or "Negative." Model training employed maxillofacial fractogram data. If two successive slices had a 95% fracture risk, the patient had a fracture. In terms of sensitivity/person for facial fractures, the recommended strategy beat the machine learning model. The recommended approach may minimize physicians' effort identifying facial bone fractures in face CT. Even though technology can't fully replace a radiologist, the recommended technique may be helpful. It reduces human error, diagnostic delays, and hospitalization costs.

Semiquantitative Approach to Amyloid Positron Emission Tomography Interpretation in Clinical Practice.

Objective Amyloid positron emission tomography (PET) plays a vital role in the in vivo detection of β-amyloid accumulation in Alzheimer's disease. Increasingly, trainees and infrequent readers are relying on semiquantitative analyses to support clinical diagnostic efforts. Our objective was to determine if the visual assessment of amyloid PET may be facilitated by relying on semiquantitative analysis. Methods We conducted a retrospective review of [ 18 F]-florbetaben PET/computed tomographies (CTs) from 2016 to 2018. Visual interpretation to determine Aβ+ status was conducted by two readers blinded to each other's interpretation. Scans were then post-processed utilizing the MIMneuro software, which generated regional-based semiquantitative Z-scores indicating cortical Aβ-burden. Results Of 167 [ 18 F]-florbetaben PET/CTs, 92/167 (reader-1) and 101/167 (reader-2) were positive for amyloid deposition (agreement = 92.2%, κ = 0.84). Additional nine scans were identified as possible Aβ-positive based solely on semiquantitative analyses. Largest semiquantitative differences were identified in the left frontal lobe (Z = 7.74 in Aβ + ; 0.50 in Aβ - ). All unilateral regions showed large statistically significant differences in Aβ-burden ( P ≤ 2.08E-28). Semiquantitative scores were highly sensitive to Aβ+ status and accurate in their ability to identify amyloid positivity, defined as a positive scan by both readers (AUC ≥ 0.90 [0.79-1.00]). Spread analyses suggested that amyloid deposition was most severe in the left posterior cingulate gyrus. The largest differences between Aβ +/Aβ- were in the left frontal lobe. Analyses using region-specific cutoffs indicated that the presence of amyloid in the temporal and anterior cingulate cortex, while exhibiting relatively low Z-scores, was most common. Conclusion Visual assessment and semiquantitative analysis provide highly congruent results, thereby enhancing reader confidence and improving scan interpretation. This is particularly relevant, given recent advances in amyloid-targeting disease-modifying therapeutics.

Development and Validation of Diagnostic Interpretation Criteria for Evaluation of F-18 Fluoromisonidazole Hypoxia PET/CT – Inter-Reader Reliability and Accuracy

<h3>Purpose/Objective(s)</h3> F-18 Fluoromisonidazole (FMISO) is a robust imaging biomarker for tumor hypoxia. Hypoxia limits effective cancer treatment, specifically radiotherapy in head and neck cancers. We have completed and have ongoing trials using hypoxia imaging to guide radiation dose de-escalation. To ensure wide applicability, we sought to develop and validate standardized qualitative and quantitative criteria to accurately and reliably interpret clinical FMISO PET/CT images. <h3>Materials/Methods</h3> All HPV+ oropharyngeal cancer patients who had FMISO PET/CT imaging before and about 2 weeks into chemoradiation therapy and pre-treatment F-18 FDG PET/CT were included. 4 independent nuclear medicine (NM) physicians with varying clinical experience of 1, 3, 5, and 15 years with no FMISO experience interpreted images. Training and reference interpretations were performed by a 5^th NM physician with 30 years of clinical and 15 years of FMISO experience. Suspicious cervical lymph nodes (LN) > 1 cm in short axis with abnormal FDG avidity on FDG PET/CT were categorized as positive or negative for FMISO uptake. A LN was considered positive if it demonstrated SUVmax visually greater than floor of mouth (FOM). FMISO SUVmax of the LN and FOM were obtained using a 3D VOI. The training and validation process consisted of: (1) initial instruction; (2) practice interpretation; (3) one on one re-training on contentious group reads; (4) final group training; (5) validation interpretation of new scans. Qualitative, visual comparison of FMISO uptake in LN to FOM, and quantitative, ratio of FMISO SUVmax of LN to FOM, were analyzed. The Fleiss κ coefficient (κ) was calculated to measure inter-reader agreement. <h3>Results</h3> 192 FMISO scans were included. Qualitative criteria mean sensitivity and specificity (SE/SP) after initial instruction were 0.773 and 0.809. κ was 0.338. Mean SE/SP after additional training were 0.976 and 0.869. κ was 0.864 [95%CI: 0.782 to 0.944]. For the quantitative criteria, ROC curve analysis was used to estimate the best threshold by maximizing the Youden index, which was determined to be SUVmax LN/SUVmax FOM > 1.2. Quantitative criteria mean SE/SP after additional training were 0.896 and 0.953. κ was 0.859 (95%CI: 0.761 to 0.944). The quantitative criteria alone demonstrated a lower sensitivity than the qualitative, 0.896 vs. 0.976. Therefore, we suggest using a hybrid method with quantitative criteria as a guide with the ultimate decision based on qualitative assessment to minimize the number of patients with false negatives and subsequent undertreatment. <h3>Conclusion</h3> The hybrid qualitative/quantitative diagnostic interpretation criteria developed for evaluating FMISO PET/CT in combination with dedicated training has substantial inter-reader agreement and excellent diagnostic performance. This suggests that multi-institutional studies that employ FMISO PET for patient stratification in prospective trials are feasible and can be implemented.

Initial Post-Therapy Imaging as a Predictor of Local Relapse after Radiotherapy in Pediatric Sarcomas

<h3>Purpose/Objective(s)</h3> To assess the value of initial post-therapy scans for predicting local relapse in patients with Ewing sarcoma (EWS) and rhabdomyosarcoma (RMS) who received radiotherapy (RT) for local control. <h3>Materials/Methods</h3> This single-center study included all patients who were diagnosed with EWS or RMS at <22 years of age, completed RT for local control and underwent post-treatment diagnostic imaging evaluations from 2007-2020. Radiologists' impressions of the primary tumor site were categorized as "positive" (residual disease within RT field), "negative" (no residual disease within RT field), or "equivocal." Each patient was assigned to one of these categories, based on the radiologists' overall impression of all off-therapy scans. The primary endpoint was local progression (LP), defined as disease progression within the RT field. Positive predictive value (PPV) and negative predictive value (NPV) quantified the ability of off-therapy imaging studies to predict LP. Actuarial rates of LP were calculated using Kaplan-Meier methods. <h3>Results</h3> Sixty-eight patients were eligible (29 EWS, 39 RMS). The median age at diagnosis was 8 years (range 0-21). All patients received chemotherapy and RT based on disease histology and risk group, per Children's Oncology Group protocols. RT was to a median dose of 50.4 Gy (range 36-55.8); 19 patients (28%) underwent surgery and RT for local control, and 49 (72%) received RT alone. Post-treatment imaging was performed 3 weeks (range <1- 13) after completion of chemotherapy and 17 weeks (range 4-52) after completion of RT. Post-treatment imaging of the primary site included PET in 51 patients (75%), MRI in 54 (79%), and CT in 5 (7%). Radiologists' overall impressions were that 1 patient's (1%) scans were positive, 50 (74%) were negative, and 17 (25%) were equivocal. 4-year LP was 100% for patients with positive scans, 12% for patients with negative scans, and 5.9% for patients with equivocal scans. The difference in the percentages of LP for patients with negative vs. equivocal scans did not reach significance (p= 0.5). Positive scans had a PPV of 100%, and negative scans an NPV of 88%. When equivocal results were considered positive (positive/equivocal vs. negative), the PPV was 11%. When equivocal results were considered negative (positive vs. equivocal/negative), the NPV was 90%. Five patients (29%) with equivocal results underwent short-term surveillance imaging at a median of 7 weeks (range 6-9) after the equivocal studies. No short-term follow-up scan demonstrated local progression. <h3>Conclusion</h3> Equivocal findings on initial post-treatment diagnostic imaging studies did not indicate a higher risk of local recurrence after RT than negative findings. These results may influence the interpretation of initial post-therapy scans and contribute to accurate counseling of patients and families.

Using Radiomics to Improve the Diagnostic Accuracy of Indeterminate Residual Primary Disease on Restaging PET/CT Imaging Following Radiation Therapy for Head and Neck Cancers

<h3>Purpose/Objective(s)</h3> Identifying residual disease (RD) in the primary site of head and neck cancer (HNC) patients treated with radiotherapy (RT) on the 3-month post RT PET/CT scans can be challenging. Radiologists have to account for RT-related inflammation versus RD. Accurate identification is key to determine the need for morbid salvage versus close observation. In this study, we hypothesized a clinical/radiomics model could complement a radiologist's interpretation in these uncertain cases. <h3>Materials/Methods</h3> We identified 182 patients with HNC treated at our institution from 2016-2020. We reviewed the 3-month post-RT PET/CT scan interpretations to obtain the radiologists' RD assessment. We classified assessments as "certain" or "uncertain." "Certain" assessments included words like "complete metabolic response," "no RD," "RD present" while "uncertain" assessments provided a differential. For uncertain cases, predictions were inferred as the most likely diagnosis ("likely radiation changes but can't rule out RD" was no RD present). If no clarification was provided, then the prediction was marked as RD present. We extracted radiomics features from the CT and PET modalities using open-source software. We gathered clinical data such as HPV status, T stage, etc. Feature reduction was performed by assessing single feature performance on a logistic regression model. Features with high correlation (>0.9) measured using Pearson coefficient were discarded. Recursive feature elimination was performed on the remaining features (only 5-10 features were allowed in the final model). We trained an explainable boosting machine model using the "certain" data and tested with "uncertain" data. Ground truth was obtained from follow-up imaging or pathology. <h3>Results</h3> Ninety-six assessments were "certain" and 86 were "uncertain." RD was present in 10 and 9 cases, respectively. The performances for the radiologists and our models are summarized in Table 1. The Clinical/CT radiomics model reduced the number of false positives from 22 to 8 at the cost of only one additional false negative. The ensemble of our CT and PET models did not improve performance. Table 1 <h3>Conclusion</h3> Even expert radiologists may struggle with identifying RD given post-RT inflammatory changes. Our Clinical/CT radiomics model greatly improved specificity with only a minor drop in sensitivity in this uncertain cohort. Machine learning models may have utility in complementing clinical experience to improve diagnostic accuracy for this challenging scenario.

Comparison of skeletal scintigraphy and standing 18 F-NaF positron emission tomography for imaging of the fetlock in 33 Thoroughbred racehorses.

This retrospective, methods comparison study aimed to compare skeletal scintigraphy and 18 F-NaF positron emission tomography (PET) for the detection of abnormalities in the fetlocks of Thoroughbred racehorses. Thirty-three horses (72 limbs) imaged with both scintigraphy and 18 F-NaF PET, for investigation of lameness or poor performance related to the fetlock, were included. Seven observers, including experienced racetrack practitioners, surgery and imaging residents, and a board-certified veterinary radiologist, independently reviewed all data for evidence of increased radiopharmaceutical uptake in 10 different regions of interest. The interobserver agreement was higher for PET (Kappa-weighted (K-w) 0.73 (0.51-0.84)) (median (range)) than for scintigraphy (0.61 (0.40-0.77)) (P<0.0001). When scintigraphy and PET were compared, the agreement was fair (K-w 0.29). More sites of increased uptake were identified using PET compared with scintigraphy. Agreement between the two modalities was higher for the palmar/plantar metacarpal/metatarsal condylar regions (K-w 0.59) than for the proximal sesamoid bones (K-w 0.25). Increased radiopharmaceutical uptake was detected in the medial proximal sesamoid bone in 6.9% and 22.2% of limbs with scintigraphy and PET, respectively. The high interobserver agreement for PET, despite the recent introduction of this technique, demonstrates the ease of clinical interpretation of PET scans. The higher number of lesions detected with PET compared with scintigraphy can be explained by the higher spatial resolution and cross-sectional nature of this modality. Study findings supported using PET in a clinical population of racehorses, in particular for the assessment of the proximal sesamoid bones.

The Potential Role of Artificial Intelligence in Lung Cancer Screening Using Low-Dose Computed Tomography.

Two large randomized controlled trials of low-dose CT (LDCT)-based lung cancer screening (LCS) in high-risk smoker populations have shown a reduction in the number of lung cancer deaths in the screening group compared to a control group. Even if various countries are currently considering the implementation of LCS programs, recurring doubts and fears persist about the potentially high false positive rates, cost-effectiveness, and the availability of radiologists for scan interpretation. Artificial intelligence (AI) can potentially increase the efficiency of LCS. The objective of this article is to review the performances of AI algorithms developed for different tasks that make up the interpretation of LCS CT scans, and to estimate how these AI algorithms may be used as a second reader. Despite the reduction in lung cancer mortality due to LCS with LDCT, many smokers die of comorbid smoking-related diseases. The identification of CT features associated with these comorbidities could increase the value of screening with minimal impact on LCS programs. Because these smoking-related conditions are not systematically assessed in current LCS programs, AI can identify individuals with evidence of previously undiagnosed cardiovascular disease, emphysema or osteoporosis and offer an opportunity for treatment and prevention.

Clinical impact of amyloid PET using 18F-florbetapir in patients with cognitive impairment and suspected Alzheimer’s disease: a multicenter study

ObjectiveAmyloid positron emission tomography (PET) can reliably detect senile plaques and fluorinated ligands are approved for clinical use. However, the clinical impact of amyloid PET imaging is still under investigation. The aim of this study was to evaluate the diagnostic impact and clinical utility in patient management of amyloid PET using 18F-florbetapir in patients with cognitive impairment and suspected Alzheimer’s disease (AD). We also aimed to determine the cutoffs for amyloid positivity for quantitative measures by investigating the agreement between quantitative and visual assessments.MethodsNinety-nine patients suspected of having AD underwent 18F-florbetapir PET at five institutions. Site-specialized physicians provided a diagnosis of AD or non-AD with a percentage estimate of their confidence and their plan for patient management in terms of medication, prescription dosage, additional diagnostic tests, and care planning both before and after receiving the amyloid imaging results. A PET image for each patient was visually assessed and dichotomously rated as either amyloid-positive or amyloid-negative by four board-certified nuclear medicine physicians. The PET images were also quantitatively analyzed using the standardized uptake value ratio (SUVR) and Centiloid (CL) scale.ResultsVisual interpretation obtained 48 positive and 51 negative PET scans. The amyloid PET results changed the AD and non-AD diagnosis in 39 of 99 patients (39.3%). The change rates of 26 of the 54 patients (48.1%) with a pre-scan AD diagnosis were significantly higher than those of 13 of the 45 patients with a pre-scan non-AD diagnosis (χ2 = 5.334, p = 0.0209). Amyloid PET results also resulted in at least one change to the patient management plan in 42 patients (42%), mainly medication (20 patients, 20%) and care planning (25 patients, 25%). Receiver-operating characteristic analysis determined the best agreement of the quantitative assessments and visual interpretation of PET scans to have an area under the curve of 0.993 at an SUVR of 1.19 and CL of 25.9.ConclusionAmyloid PET using 18F-florbetapir PET had a substantial clinical impact on AD and non-AD diagnosis and on patient management by enhancing diagnostic confidence. In addition, the quantitative measures may improve the visual interpretation of amyloid positivity.