Improving sensitivity through data augmentation with synthetic lymph node metastases for AI-based analysis of PSMA PET-CT images.

BACKGROUND: We developed a fully automated artificial intelligence (AI)AI-based-based method for detecting suspected lymph node metastases in prostate-specific membrane antigen (PSMA)(PSMA) positron emission tomography-computed tomography (PET-CT)(PET-CT) images of prostate cancer patients by using data augmentation that adds synthetic lymph node metastases to the images to expand the training set. METHODS: Synthetic data were derived from original training images to which new synthetic lymph node metastases were added. Thus, the original training set from a previous study (n = 420) was expanded by one synthetic image for every original image (n = 840), which was used to train an AI model. The performance of the AI model was compared to that of nuclear medicine physicians and a previously developed AI model. The human readers were alternately used as a reference and compared to either another reading or AI model. RESULTS: The new AI model had an average sensitivity of 84% for detecting lymph node metastases compared with 78% for human readings. Our previously developed AI method without synthetic data had an average sensitivity of 79%. The number of false positive lesions were slightly higher for the new AI model (average 3.3 instances per patient) compared to human readings and the previous AI model (average 2.8 instances per patient), while the number of false negative lesions was lower. CONCLUSIONS: Creating synthetic lymph node metastases, as a form of data augmentation, on [18F]PSMA-1007F]PSMA-1007 PETPET-CT-CT images improved the sensitivity of an AI model for detecting suspected lymph node metastases. However, the number of false positive lesions increased somewhat.

MRI morphological characteristics of lymph nodes in anal squamous cell carcinoma.

OBJECTIVES: Pre-treatment staging of anal squamous cell carcinoma (ASCC) includes pelvic MRI and [18F]-fluorodeoxyglucose positron emission tomography with computed tomography (PET-CT). MRI criteria to define lymph node metastases (LNMs) in ASCC are currently lacking. The aim of this study was to describe the morphological characteristics of lymph nodes (LNs) on MRI in ASCC patients with PET-CT-positive LNs. METHODS: ASCC patients treated at Skåne University Hospital between 2009 and 2017 were eligible for inclusion if at least one positive LN according to PET-CT and a pre-treatment MRI were present. All PET-CT-positive LNs and PET-CT-negative LNs were retrospectively identified on baseline MRI. Each LN was independently classified according to pre-determined morphological characteristics by two radiologists blinded to clinical patient information. RESULTS: Sixty-seven ASCC patients were included, with a total of 181 PET-CT-positive LNs identified on baseline MRI with a median short-axis diameter of 9.0 mm (range 7.5-12 mm). MRI morphological characteristics of PET-CT-positive LNs included regular contour (87%), round shape (89%), and homogeneous signal intensity on T2-weighed images (67%). An additional 78 PET-CT-negative LNs were identified on MRI. These 78 LNs had a median size of 6.8 mm (range 5.5-8.0 mm). The majority of PET-CT-negative LNs had a regular contour, round shape, and a homogeneous signal that was congruent to the primary tumor. CONCLUSIONS: There are MRI-specific morphological characteristics for pelvic LNs in ASCC. PET-CT-positive and negative LNs share similar morphological features apart from size, with PET-CT-positive LNs being significantly larger. Further studies are needed to determine discrimination criteria for LNM in ASCC.

A novel model of artificial intelligence based automated image analysis of CT urography to identify bladder cancer in patients investigated for macroscopic hematuria.

OBJECTIVE: To evaluate whether artificial intelligence (AI) based automatic image analysis utilising convolutional neural networks (CNNs) can be used to evaluate computed tomography urography (CTU) for the presence of urinary bladder cancer (UBC) in patients with macroscopic hematuria. METHODS: Our study included patients who had undergone evaluation for macroscopic hematuria. A CNN-based AI model was trained and validated on the CTUs included in the study on a dedicated research platform (Recomia.org). Sensitivity and specificity were calculated to assess the performance of the AI model. Cystoscopy findings were used as the reference method. RESULTS: The training cohort comprised a total of 530 patients. Following the optimisation process, we developed the last version of our AI model. Subsequently, we utilised the model in the validation cohort which included an additional 400 patients (including 239 patients with UBC). The AI model had a sensitivity of 0.83 (95% confidence intervals [CI], 0.76-0.89), specificity of 0.76 (95% CI 0.67-0.84), and a negative predictive value (NPV) of 0.97 (95% CI 0.95-0.98). The majority of tumours in the false negative group (n = 24) were solitary (67%) and smaller than 1 cm (50%), with the majority of patients having cTaG1-2 (71%). CONCLUSIONS: We developed and tested an AI model for automatic image analysis of CTUs to detect UBC in patients with macroscopic hematuria. This model showed promising results with a high detection rate and excessive NPV. Further developments could lead to a decreased need for invasive investigations and prioritising patients with serious tumours.

AI-based fully automatic image analysis: Optimal abdominal and thoracic segmentation volumes for estimating total muscle volume on computed tomography scans.

Objectives: Evaluation of sarcopenia from computed tomography (CT) is often based on measuring skeletal muscle area on a single transverse slice. Automatic segmentation of muscle volume has a lower variance and may be a better proxy for the total muscle volume than single-slice areas. The aim of the study was to determine which abdominal and thoracic anatomical volumes were best at predicting the total muscle volume. Methods: A cloud-based artificial intelligence tool (recomia.org) was used to segment all skeletal muscle of the torso of 994 patients who had performed whole-torso CT 2008-2020 for various clinical indications. Linear regression models for several anatomical volumes and single-slice areas were compared with regard to predicting the total torso muscle volume. Results: The muscle volume from the tip of the coccyx and 25 cm cranially was the best of the abdominal volumes and was significantly better than the L3 slice muscle area (R2 0.935 vs 0.830, P < 0.0001). For thoracic volumes, the muscle volume between the top of the sternum to the lower bound of the Th12 vertebra showed the best correlation with the total volume, significantly better than the Th12 slice muscle area (R2 0.892 vs 0.775, P < 0.0001). Adjusting for body height improved the correlation slightly for all measurements but did not significantly change the ordering. Conclusions: We identified muscle volumes that can be reliably segmented by automated image analysis which is superior to single slice areas in predicting total muscle volume.

Applications of Artificial Intelligence in PSMA PET/CT for Prostate Cancer Imaging.

Prostate-specific membrane antigen (PSMA) positron emission tomography/computed tomography (PET/CT) has emerged as an important imaging technique for prostate cancer. The use of PSMA PET/CT is rapidly increasing, while the number of nuclear medicine physicians and radiologists to interpret these scans is limited. Additionally, there is variability in interpretation among readers. Artificial intelligence techniques, including traditional machine learning and deep learning algorithms, are being used to address these challenges and provide additional insights from the images. The aim of this scoping review was to summarize the available research on the development and applications of AI in PSMA PET/CT for prostate cancer imaging. A systematic literature search was performed in PubMed, Embase and Cinahl according to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines. A total of 26 publications were included in the synthesis. The included studies focus on different aspects of artificial intelligence in PSMA PET/CT, including detection of primary tumor, local recurrence and metastatic lesions, lesion classification, tumor quantification and prediction/prognostication. Several studies show similar performances of artificial intelligence algorithms compared to human interpretation. Few artificial intelligence tools are approved for use in clinical practice. Major limitations include the lack of external validation and prospective design. Demonstrating the clinical impact and utility of artificial intelligence tools is crucial for their adoption in healthcare settings. To take the next step towards a clinically valuable artificial intelligence tool that provides quantitative data, independent validation studies are needed across institutions and equipment to ensure robustness.

Retrospective evaluation of the predictive value of tumour burden at baseline [68 Ga]Ga-DOTA-TOC or -TATE PET/CT and tumour dosimetry in GEP-NET patients treated with PRRT.

PURPOSE: There is a lack of validated imaging biomarkers for prediction of response to peptide receptor radionuclide therapy (PRRT). The primary objective was to evaluate if tumour burden at baseline PET/CT could predict treatment outcomes to PRRT with [177Lu]Lu-DOTA-TATE. Secondary objectives were to evaluate if there was a correlation between tumour burden and mean tumour absorbed dose (AD) during first cycle, and if mean tumour AD or the relative change of tumour burden at first follow-up PET/CT could predict progression free survival (PFS) or overall survival (OS). METHODS: Patients with gastroenteropancreatic neuroendocrine tumour (GEP-NET) treated with [177Lu]Lu-DOTA-TATE PRRT were retrospectively included. Tumour burden was quantified from [68 Ga]Ga-DOTA-TOC/TATE PET/CT-images at baseline and first follow-up and expressed as; whole-body somatostatin receptor expressing tumour volume (SRETVwb), total lesion somatostatin receptor expression (TLSREwb), largest tumour lesion diameter and highest SUVmax. The relative change of tumour burden was evaluated in three categories. Mean tumour AD was estimated from the first cycle of PRRT. PFS was defined as time from start of PRRT to radiological or clinical progression. OS was evaluated as time to death. Kaplan Meier survival curves and log-rank test were used to compare PFS and OS between different groups. RESULTS: Thirty-one patients had a baseline PET/CT < 6 months before treatment and 25 had a follow-up examination. Median tumour burden was 132 ml (IQR 61-302) at baseline and 71 ml (IQR 36-278) at follow-up. Twenty-two patients had disease progression (median time to progression 17.2 months) and 9 patients had no disease progression (median follow-up 28.7 months). SRETVwb dichotomized by the median at baseline was not associated with longer PFS (p = 0.861) or OS (p = 0.937). Neither TLSREwb, largest tumour lesion or SUVmax showed significant predictive value. There was a moderately strong correlation, however, between SUVmax and mean tumour AD r = 0.705, p < 0.001, but no significant correlation between SRETVwb nor TLSREwb and mean tumour AD. An increase of SRETVwb, TLSREwb or largest tumour lesion at first follow-up PET/CT was significantly correlated with shorter PFS/OS. CONCLUSION: Tumour burden at baseline showed no predictive value of PFS/OS after PRRT in this small retrospective study. An increase of tumour burden was predictive of worse outcome.

Artificial Intelligence-Based Organ Delineation for Radiation Treatment Planning of Prostate Cancer on Computed Tomography.

Purpose: Meticulous manual delineations of the prostate and the surrounding organs at risk are necessary for prostate cancer radiation therapy to avoid side effects to the latter. This process is time consuming and hampered by inter- and intraobserver variability, all of which could be alleviated by artificial intelligence (AI). This study aimed to evaluate the performance of AI compared with manual organ delineations on computed tomography (CT) scans for radiation treatment planning. Methods and Materials: Manual delineations of the prostate, urinary bladder, and rectum of 1530 patients with prostate cancer who received curative radiation therapy from 2006 to 2018 were included. Approximately 50% of those CT scans were used as a training set, 25% as a validation set, and 25% as a test set. Patients with hip prostheses were excluded because of metal artifacts. After training and fine-tuning with the validation set, automated delineations of the prostate and organs at risk were obtained for the test set. Sørensen-Dice similarity coefficient, mean surface distance, and Hausdorff distance were used to evaluate the agreement between the manual and automated delineations. Results: The median Sørensen-Dice similarity coefficient between the manual and AI delineations was 0.82, 0.95, and 0.88 for the prostate, urinary bladder, and rectum, respectively. The median mean surface distance and Hausdorff distance were 1.7 and 9.2 mm for the prostate, 0.7 and 6.7 mm for the urinary bladder, and 1.1 and 13.5 mm for the rectum, respectively. Conclusions: Automated CT-based organ delineation for prostate cancer radiation treatment planning is feasible and shows good agreement with manually performed contouring.