Can the Diagnostic Accuracy of Bone Scintigraphy be Maintained with Half the Scanning Time?

Keeping Up With ChatGPT: Evaluating Its Recognition and Interpretation of Nuclear Medicine Images

PURPOSE

The latest iteration of GPT4 (generative pretrained transformer) is a large multimodal model that can integrate both text and image input, but its performance with medical images has not been systematically evaluated. We studied whether ChatGPT with GPT-4V(ision) can recognize images from common nuclear medicine examinations and interpret them.

PATIENTS AND METHODS

Fifteen representative images (scintigraphy, 11; PET, 4) were submitted to ChatGPT with GPT-4V(ision), both in its Default and "Advanced Data Analysis (beta)" version. ChatGPT was asked to name the type of examination and tracer, explain the findings and whether there are abnormalities. ChatGPT should also mark anatomical structures or pathological findings. The appropriateness of the responses was rated by 3 nuclear medicine physicians.

RESULTS

The Default version identified the examination and the tracer correctly in the majority of the 15 cases (60% or 53%) and gave an "appropriate" description of the findings or abnormalities in 47% or 33% of cases, respectively. The Default version cannot manipulate images. "Advanced Data Analysis (beta)" failed in all tasks in >90% of cases. A "major" or "incompatible" inconsistency between 3 trials of the same prompt was observed in 73% (Default version) or 87% of cases ("Advanced Data Analysis (beta)" version).

CONCLUSIONS

Although GPT-4V(ision) demonstrates preliminary capabilities in analyzing nuclear medicine images, it exhibits significant limitations, particularly in its reliability (ie, correctness, predictability, and consistency).

The use of MRI for the imaging of metastatic bone lesions

Skeletal metastatic disease accounts for significant overall morbidity in cancer patients. Accurate and accessible imaging forms an integral part of the investigation for patients with suspected or known skeletal metastatic disease; it is considered indispensable in making appropriate oncological treatment decisions. Magnetic resonance imaging (MRI) is a contemporary imaging modality that provides excellent spatial and contrast resolution for bone and soft tissues. Therefore, it is particularly useful for imaging patients suffering from metastatic skeletal disease. This review provides a fundamental overview of the physics and image generation of MRI. The most commonly used MRI sequences in the investigation of metastatic skeletal disease are also discussed. Additionally, a review of the pathophysiological basis of metastatic bone disease is presented, along with an introduction to the interpretation of MRI sequences obtained for metastatic bone disease. Finally, the strengths and drawbacks of MRI are considered in comparison to alternative imaging modalities for the investigation of this common and important oncological complication.

Radiopharmaceutical extravasation in bone scintigraphy: a cross-sectional study

OBJETIVES

Tc-99m Hydroxymethylene diphosphonate (HMDP) bone scintigraphy is commonly used to diagnose bone disorders. We aimed to quantify and characterize the occurrence of radiopharmaceutical extravasation in bone scintigraphy, using Tc-99m HMDP, as well as to compare the visual classification of the events with an independent analysis using image processing software.

METHODS

We conducted a cross-sectional study, using data from a total of 400 (9.1%) exams, randomly selected from all the procedures performed in 2018 in the Portuguese Institute of Oncology of Porto, Portugal. Prevalence estimate and the corresponding 95% confidence interval (CI) was computed for the presence of extravasation. Odds ratios and 95% CI were computed to quantify the association between demographic and clinical characteristics, and the occurrence of extravasation.

RESULTS

The prevalence of Tc-99m HMDP extravasation was 26.5% (95% CI: 22.4-31.0). Those from an inpatient setting had almost seven-fold higher odds of extravasation than those from an outpatient setting. When the wrist was used for administration, there was three times more odds of extravasation when compared to the use of hand. There were statistically significant differences in the median scores of extravasations severity obtained from image processing software according to the different grades attributed by visual appreciation ( P  < 0.001).

CONCLUSION

Tc-99m HMDP extravasation occurred in one out of four patients, being more frequent among those from an inpatient setting and when the wrist was used for administration. Visual appreciation of the extravasation seems to be acceptable to classify its severity.