Considerations on Baseline Generation for Imaging AI Studies Illustrated on the CT-Based Prediction of Empyema and Outcome Assessment

The role of artificial intelligence in the diagnosis, imaging, and treatment of thoracic empyema

PURPOSE OF REVIEW

The management of thoracic empyema is often complicated by diagnostic delays, recurrence, treatment failures and infections with antibiotic resistant bacteria. The emergence of artificial intelligence (AI) in healthcare, particularly in clinical decision support, imaging, and diagnostic microbiology raises great expectations in addressing these challenges.

RECENT FINDINGS

Machine learning (ML) and AI models have been applied to CT scans and chest X-rays to identify and classify pleural effusions and empyema with greater accuracy. AI-based analyses can identify complex imaging features that are often missed by the human eye, improving diagnostic precision. AI-driven decision-support algorithms could reduce time to diagnosis, improve antibiotic stewardship, and enhance more precise and less invasive surgical therapy, significantly improving clinical outcomes and reducing inpatient hospital stays.

SUMMARY

ML and AI can analyse large datasets and recognize complex patterns and thus have the potential to enhance diagnostic accuracy, preop planning for thoracic surgery, and optimize surgical treatment strategies, antibiotic therapy, antibiotic stewardship, monitoring complications, and long-term patient management outcomes.

Time Is Money: Considerations for Measuring the Radiological Reading Time

Timestamps in the Radiology Information System (RIS) are a readily available and valuable source of information with increasing significance, among others, due to the current focus on the clinical impact of artificial intelligence applications. We aimed to evaluate timestamp-based radiological dictation time, introduce timestamp modeling techniques, and compare those with prospective measured reporting. Dictation time was calculated from RIS timestamps between 05/2010 and 01/2021 at our institution (n = 108,310). We minimized contextual outliers by simulating the raw data by iteration (1000, vector size (µ/sd/λ) = 100/loop), assuming normally distributed reporting times. In addition, 329 reporting times were prospectively measured by two radiologists (1 and 4 years of experience). Altogether, 106,127 of 108,310 exams were included after simulation, with a mean dictation time of 16.62 min. Mean dictation time was 16.05 min head CT (44,743/45,596), 15.84 min for chest CT (32,797/33,381), 17.92 min for abdominal CT (n = 22,805/23,483), 10.96 min for CT foot (n = 937/958), 9.14 min for lumbar spine (881/892), 8.83 min for shoulder (409/436), 8.83 min for CT wrist (1201/1322), and 39.20 min for a polytrauma patient (2127/2242), without a significant difference to the prospective reporting times. In conclusion, timestamp analysis is useful to measure current reporting practice, whereas body-region and radiological experience are confounders. This could aid in cost–benefit assessments of workflow changes (e.g., AI implementation).