The bone scan index (BSI), initially introduced for metastatic prostate cancer, quantifies the osseous tumor load from planar bone scans. Following the basic idea of radiomics, this method incorporates specific deep-learning techniques (artificial neural network) in its development to provide automatic calculation, feature extraction, and diagnostic support. As its performance in tumor entities, not including prostate cancer, remains unclear, our aim was to obtain more data about this aspect. The results of BSI evaluation of bone scans from 951 consecutive patients with different tumors were retrospectively compared to clinical reports (bone metastases, yes/no). Statistical analysis included entity-specific receiver operating characteristics to determine optimized BSI cut-off values. In addition to prostate cancer (cut-off = 0.27%, sensitivity (SN) = 87%, specificity (SP) = 99%), the algorithm used provided comparable results for breast cancer (cut-off 0.18%, SN = 83%, SP = 87%) and colorectal cancer (cut-off = 0.10%, SN = 100%, SP = 90%). Worse performance was observed for lung cancer (cut-off = 0.06%, SN = 63%, SP = 70%) and renal cell carcinoma (cut-off = 0.30%, SN = 75%, SP = 84%). The algorithm did not perform satisfactorily in melanoma (SN = 60%). For most entities, a high negative predictive value (NPV ≥ 87.5%, melanoma 80%) was determined, whereas positive predictive value (PPV) was clinically not applicable. Automatically determined BSI showed good sensitivity and specificity in prostate cancer and various other entities. Particularly, the high NPV encourages applying BSI as a tool for computer-aided diagnostic in various tumor entities.

While multi-parametric magnetic resonance imaging (mpMRI) is known to be a specific and reliable modality for the diagnosis of non-metastatic prostate cancer (PC), positron emission tomography (PET) using ⁶⁸Ga labeled ligands targeting the prostate-specific membrane antigen (PSMA) is known for its reliable detection of prostate cancer, being the most sensitive modality for the assessment of the extra-prostatic extension of the disease and the establishment of a diagnosis, even before biopsy. Background/Objectives: Here, we compared these modalities in regards to the localization of intraprostatic cancer lesions prior to local HDR brachytherapy. Methods: A cohort of 27 patients received both mpMRI and PSMA-PET/CT. Based on 24 intraprostatic segments, two readers each scored the risk of tumor-like alteration in each imaging modality. The detectability was evaluated using receiver operating characteristic (ROC) analysis. The histopathological findings from biopsy were used as the gold standard in each segment. In addition, we applied a patient-based “congruence” concept to quantify the interobserver and intermodality agreement. Results: For the ROC analysis, we included 447 segments (19 patients), with their respective histological references. The two readers of the MRI reached an AUC of 0.770 and 0.781, respectively, with no significant difference (p = 0.75). The PET/CT readers reached an AUC of 0.684 and 0.608, respectively, with a significant difference (p < 0.001). The segment-wise intermodality comparison showed a significant superiority of MRI (AUC = 0.815) compared to PET/CT (AUC = 0.690) (p = 0.006). Via a patient-based analysis, a superiority of MRI in terms of relative agreement with the biopsy result was observed (n = 19 patients). We found congruence scores of 83% (MRI) and 76% (PET/CT, p = 0.034), respectively. Using an adjusted “near total agreement” score (adjacent segments with positive scores of 4 or 5 counted as congruent), we found an increase in the agreement, with a score of 96.5% for MRI and 92.7% for PET/CT, with significant difference (p = 0.024). Conclusions: This study suggests that in a small collective of low-/intermediate risk prostate cancer, mpMRI is superior for the detection of intraprostatic lesions as compared to PSMA-PET/CT. We also found a higher relative agreement between MRI and biopsy as compared to that for PET/CT. However, further studies including a larger number of patients and readers are necessary to draw solid conclusions.

Objectives: This study evaluated 3 LLM chatbots (GPT-3.5-turbo, GPT-4-turbo, and GPT-4o) on their effectiveness in supporting patient education by answering common patient questions for CT, MRI, and DSA informed consent, assessing their accuracy and clarity. Methods: Two radiologists formulated 90 questions categorized as general, clinical, or technical. Each LLM answered every question 5×. Radiologists then rated the responses for medical accuracy and clarity, while medical physicists assessed technical accuracy using a Likert scale. Semantic similarity was analyzed with SBERT and cosine similarity. Results: Ratings improved with newer model versions. Linear mixed-effects models revealed that GPT-4 models were rated significantly higher than GPT-3.5 (P < .001) by both physicians and physicists. However, physicians’ ratings for GPT-4 models showed a significant performance decrease for complex modalities like DSA and MRI (P < .01), a pattern not observed in physicists’ ratings. SBERT analysis revealed high internal consistency across all models. SBERT analysis revealed high internal consistency across all models. Conclusion: Variability in ratings revealed that while models effectively handled general and technical questions, they struggled with contextually complex medical inquiries requiring personalized responses and nuanced understanding. Statistical analysis confirms that while newer models are superior, their performance is modality-dependent and perceived differently by clinical and technical experts. Advances in knowledge: This study evaluates the potential of LLMs to enhance informed consent in radiology, highlighting strengths in general and technical questions while noting limitations with complex clinical inquiries, with performance varying significantly by model type and imaging modality.