TSThomas Saliba

Thomas Saliba

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article


Thomas Saliba has a total of 91 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    Trust me, I’m a chatbot: The rise of vibe medicine in healthcare

    by Thomas Saliba, Guillaume Fahrni, Rohin Francis

  • Open Access

    Comparing Virtual Reality Headsets With Standard Diagnostic Workstations for MRI Grading of Knee Meniscus Lesions: A Pilot Study

    by Thomas Saliba, Sanjiva Pather

    Background: Virtual reality (VR) is becoming increasingly popular in radiology, with a variety of diagnostic applications being tested. However, VR is yet to be tested in the context of MRI studies. This study evaluated the agreement between a VR headset and a standard radiology workstation in grading meniscus lesions on knee MRI. Methods: Fifty‐two knee MRIs were reviewed by a consultant musculoskeletal radiologist and a radiology resident. Each interpreted the examinations on a diagnostic screen and then re‐evaluated them using a VR headset. Meniscus lesions were graded using Lotysch's five‐segment classification. Agreement was assessed using Cohen's kappa and descriptive statistics, with significance at p < 0.05. Results: Cohen's kappa was not always calculable due to asymmetry in data. The consultant showed moderate‐to‐near‐perfect agreement between VR and standard screens, 76% for the medial and 87% for the lateral meniscus. The resident showed moderate to minor agreement, 64% for the medial and 70% for the lateral meniscus. Agreement for nonpathological segments was high: 94% (consultant) and 76% (resident). Agreement dropped for pathological segments, particularly type 2b superior lesions (29% consultant and 0% resident). Conclusions: VR headsets demonstrate strong comparability for excluding meniscal lesions and moderate accuracy in grading pathology, although agreement for pathological segments was limited. Although not a replacement for diagnostic screens, they may be useful in select cases. As technology advances, VR could support remote interpretation and supplement traditional radiology workflows.

  • Case 355

    by Thomas Saliba, Vincent Dunet

    History: An 8-year-old girl presented to the pediatrician with headaches that she had been experiencing for the last 2–3 years but that had become intolerable in the previous 8 days. The headaches were described by the patient as not having any lateralization and mainly involving the frontal and occipital regions. The headaches were rated by the patient as being an eight out of 10 in terms of pain, which prevented her from sleeping. Analgesic therapy proved to be ineffective. The patient had no photophobia or phonophobia but had increased fatigue alongside developing nausea without vomiting. The patient did not have any dysphagia, ataxia, vertigo, or elocution issues. Previous MRI performed in a provincial hospital showed a mass in the posterior fossa that was displacing the brainstem. However, being unable to get an appointment to see a neurosurgeon in a timely manner, the patient presented to the emergency department of our university hospital due to worsening symptoms. In order to exclude an intracranial origin of the symptoms, brain MRI with contrast enhancement was performed ( Figure ).

  • Open Access

    Virtual reality for managing pain and fear and anxiety during pediatric needle procedures: an umbrella review

    by Thomas Saliba, Guillaume Fahrni

    Needle-involving procedures often cause significant pain and anxiety, particularly in the pediatric population. Virtual reality (VR) has emerged as a nonpharmacological tool for alleviating these effects. However, the methodological quality and consistency of current meta-analyses of VR tools remain unclear. Here we performed an umbrella review to evaluate published meta-analyses that assessed the ability of VR to reduce pain, fear, and anxiety during medical procedures involving needles. A systematic search following the PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-analyses) 2020 guidelines was performed of the PubMed, Web of Science, Embase, ScienceDirect, and Cochrane Library databases in May 2026. Study methodological quality was assessed using A MeaSurement Tool to Assess systematic Reviews, version 2 (AMSTAR 2). The primary study overlap was quantified using the corrected covered area (CCA) and visualized using the GROOVE (Graphical Representation of Overlap of OVErviews) approach. An umbrella meta-analysis was performed using a metaumbrella analysis tool, with sensitivity analyses varying for inter- and intrastudy variance estimators and correlations. Eleven meta-analyses were included, containing 49 unique primary studies. The CCA was 15.7%, indicating very high overlap. The AMSTAR 2 ratings were predominantly low or critically low, with only one high-confidence review. The umbrella meta-analysis demonstrated statistically significant reductions in pain (I²=90.5%), fear (I²=91.1%), and anxiety (I²=89.9%), with a significant publication bias for pain and fear. The GRADE (Grading of Recommendations Assessment, Development and Evaluation) classification was "very weak" for all 3 outcomes. Although available meta-analyses consistently suggest that VR reduces procedural pain, fear, and anxiety in children, the evidence base is undermined by high heterogeneity, substantial primary study overlap, and predominantly low methodological quality. VR may serve as an additional distraction tool for selected pediatric needle procedures; however, current evidence is insufficient to recommend it as a replacement for established analgesic strategies.

  • Open Access

    Age-Related Distribution of the Copper-Beaten Skull Appearance on Pediatric Cephalometric Radiographs: A Retrospective Single-Center Study

    by Thomas Saliba, Alessandro De Leucio, Paolo Simoni, Grammatina Boitsios

    Background: The copper-beaten skull (CBS) appearance on lateral skull radiographs has historically been associated with raised intracranial pressure, but convolutional markings may also occur during normal cranial development. We aimed to characterize the age-related distribution of CBS in a contemporary pediatric orthodontic referral cohort and to compare the observed pattern with classic historical series. Methods: This single-center retrospective study included 748 lateral skull radiographs from unique patients aged 3–19 years obtained for orthodontic indications between July 2008 and April 2023 (378 female, 370 male). Radiographs were graded 0–3 by two blinded radiologists, with disagreements resolved by consensus. Because the outcome was ordinal and the contingency table contained sparse cells, analyses included a Monte Carlo permutation test, Spearman and Kendall rank correlations, proportional-odds ordinal logistic regression with age, quadratic age, sex, age-by-sex interaction, and sampling period, plus bootstrap and threshold-specific robustness analyses. Results: Grade distribution was grade 0, n = 55 (7.4%); grade 1, n = 162 (21.7%); grade 2, n = 444 (59.4%); and grade 3, n = 87 (11.6%). Monte Carlo testing confirmed an age–grade association (p = .00055), although rank correlations were weak (Spearman’s ρ = −0.149; Kendall’s τ-b = −0.115; both p < .001). The ordinal model showed a significant nonlinear age effect (linear OR = 0.913, 95% CI 0.858–0.971, p = .004; quadratic OR = 0.977, 95% CI 0.964–0.991, p = .001). Sex, age-by-sex interaction, and sampling period were not significant. Results were directionally similar in the date-restricted principal-period sensitivity cohort (n = 554). Conclusion: Convolutional skull markings were common in this pediatric orthodontic referral cohort and varied with age but not significantly with sex. A copper-beaten skull appearance should not be interpreted in isolation as evidence of raised intracranial pressure. Because the cohort was selectively supplemented by age and participants were not systematically screened for neurological or craniofacial disease, the observed frequency is not a population prevalence estimate and does not establish physiological normality.

  • Open Access

    Incidentally detected double-chambered right ventricle in an adult

    by Thomas Saliba, David Rotzinger, Guillaume Fahrni, Philippe Thoma

  • Open Access

    Large Language Model-Generated Differential Diagnoses in Radiology Education: Comparison with a Standard Casebook

    by Pauline Chapellier, Jacopo Ferrari, Thomas Saliba, Patrick Jeltsch, Mustafa Mohamed, Sofyan Jankovski, Gorun Ilanjian, Marta Epis, Virginia Pansini, Federica Bragaglia, et al.

    Background/Objectives: Large language models (LLMs) are increasingly explored for radiology education, but their role in differential diagnosis learning remains under-investigated. This study evaluates the perceived usefulness of LLM-generated differential diagnoses compared with a standard radiology casebook. Methods: In this multi-center study, radiology trainees at junior (years 1–2) and advanced (years 3–5) levels evaluated 225 cases from a gold-standard casebook spanning nine subspecialties. Participants ranked the usefulness of their personal clinical experience, the casebook, and LLM teaching, and rated the LLM output using a five-point Likert scale across Clarity, Trust, Differential Usefulness, and Diagnostic Usefulness. Results: Thirteen trainees (4 junior, 9 advanced) completed 2425 evaluations. Overall, the casebook was rated most useful (mean rank 1.7 ± 0.2), followed by LLM teaching (1.8 ± 0.3) and personal experience (2.4 ± 0.2; p = 0.023), with no significant difference between LLM and Textbook (p = 0.438). Junior trainees favored LLM teaching more than advanced trainees (first-rank 66.6% vs. 22.1%; p = 0.037). Across subspecialties, the casebook consistently ranked highest, with LLM slightly lower and experience lowest. LLM teaching received high ratings for Clarity (4.4 ± 0.3), Trust (4.3 ± 0.3), Differential Usefulness (4.3 ± 0.4), and Diagnostic Usefulness (4.2 ± 0.4), with no statistically significant difference between domains (p = 0.149). Conclusions: LLM-generated differential diagnoses are clear, trustworthy, and perceived as highly useful for education, nearing the perceived value of a standard casebook, especially for junior trainees. While textbooks remain essential, LLMs hold promise as supplementary tools, but caution is needed due to potential inaccuracies and their inability to replicate image-based teaching.


  • Open Access

    Giant intra-cardiac undifferentiated pleomorphic sarcoma: a case report

    by Thomas Saliba, Anna Nowacka, Panagiotis Antiochos, David Rotzinger

    Background: Primary cardiac tumours are rare. Though most are benign, they significantly interfere with cardiac function. Malignant tumours, predominantly sarcomas, include the highly aggressive undifferentiated pleomorphic cardiac sarcoma that often manifests with non-specific symptoms. Case Summary: A 72-year-old man presented with weeks-long fatigue, appetite loss, and chronic dyspnoea (NYHA III) following a previous acute coronary syndrome and occupational exposure to toxic solvents. His examination revealed dyspnoea and elevated blood pressure, while imaging uncovered a large intracavitary mass at the junction of the inferior vena cava and right atrium. Multiple metastatic lesions to the pleura, lung, pericardium, and lymph nodes were identified on CT, PET-CT, and MRI, with ultrasound confirming significant tricuspid stenosis caused by the mass. Histopathological analysis of a pleural metastasis established the diagnosis of an undifferentiated pleomorphic sarcoma. Discussion: Comprehensive imaging modalities (CT, PET-CT, MRI, ultrasound) are essential for characterizing cardiac masses and staging malignancy. Although surgical resection is the primary treatment, palliative chemotherapy with agents such as pazopanib, carboplatin, gemcitabine, and docetaxel is employed due to widespread metastases and poor prognosis. Undifferentiated pleomorphic cardiac sarcomas remain diagnostically challenging, with treatment focused on symptom relief and palliative care in advanced disease. Early detection remains vital for quality of life.

  • Open Access

    Can an Unenhanced Reduced-Dose ECG-Gated CT of the Aorta Replace an ECG-Gated CT-Angiography for Diameter Follow-Up of the Ascending Aorta?

    by Thomas Saliba, Denis Tack, Nicolas Naccarella, Sanjiva Pather, David Rotzinger, Olivier Cappeliez

    Electrocardiogram (ECG)-gated contrast-enhanced computed tomography angiography (CTA) is the reference method for follow-up of ascending aortic aneurysms but delivers substantially higher radiation doses than ECG-gated non-contrast CT (NCCT). NCCT can be acquired at a lower dose while enabling measurements of the aortic outer diameter. This study aimed to quantify the radiation dose of both techniques and determine whether a significant difference exists in ascending thoracic aorta diameter measurements between NCCT and CTA. Eighty patients who underwent ECG-gated cardiac CT for suspected coronary artery disease were retrospectively analyzed. Three observers measured the ascending aortic diameter at the level of the pulmonary artery in a plane perpendicular to the aorta on both NCCT and CTA images. Inter-rater reliability was assessed using intraclass correlation coefficients, and paired samples t-tests were used to evaluate measurement differences. Dose-length products (DLP) were collected. Median DLP values were 16.1 mGy·cm (interquartile range 11.8–25.1) for NCCT and 190.3 mGy·cm (interquartile range 120.5–298.9) for CTA. NCCT measurements were consistently larger than CTA measurements, with mean differences of 2.1 ± 0.8 mm, 2.6 ± 0.96 mm, and 2.9 ± 1.09 mm for the senior radiologist, junior radiologist, and resident, respectively (all p < 0.001). Inter-observer agreement was excellent (ICC = 0.99, p < 0.001). NCCT delivered an 11.8-fold lower radiation dose than CTA. NCCT may replace CTA for ascending aortic diameter follow-up if measurements are adjusted by approximately 2–3 mm relative to CTA-derived inner-diameter thresholds.