Kameel Khabaz

Mr. Kameel Khabaz

orcid

Research Focus & Metrics

Academic Career Length

6Years (2021-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
29 October 2024 in Journal of Surgical Oncology
Verified
Journal Article
Open access
KKKameel Khabaz
Kameel Khabaz
NNNicole J. Newman‐Hung
Nicole J. Newman‐Hung
JKJennifer R. Kallini
Jennifer R. Kallini
JKJoseph Kendal
Joseph Kendal
ACAlexander B. Christ
Alexander B. Christ
LWLauren E. Wessel
Lauren E. Wessel
Background and Objectives: The potential impacts of artificial intelligence (AI) chatbots on care for patients with bone sarcoma is poorly understood. Elucidating potential risks and benefits would allow surgeons to define appropriate roles for these tools in clinical care. Methods: Eleven questions on bone sarcoma diagnosis, treatment, and recovery were inputted into three AI chatbots. Answers were assessed on a 5‐point Likert scale for five clinical accuracy metrics: relevance to the question, balance and lack of bias, basis on established data, factual accuracy, and completeness in scope. Responses were quantitatively assessed for empathy and readability. The Patient Education Materials Assessment Tool (PEMAT) was assessed for understandability and actionability. Results: Chatbots scored highly on relevance (4.24) and balance/lack of bias (4.09) but lower on basing responses on established data (3.77), completeness (3.68), and factual accuracy (3.66). Responses generally scored well on understandability (84.30%), while actionability scores were low for questions on treatment (64.58%) and recovery (60.64%). GPT‐4 exhibited the highest empathy (4.12). Readability scores averaged between 10.28 for diagnosis questions to 11.65 for recovery questions. Conclusions: While AI chatbots are promising tools, current limitations in factual accuracy and completeness, as well as concerns of inaccessibility to populations with lower health literacy, may significantly limit their clinical utility.

30 November 2025 in International Journal of Medical Informatics
Verified
Journal Article
Open access
KKKameel Khabaz
Kameel Khabaz
ZPZaid Parekh
Zaid Parekh
SSShayan Saeed
Shayan Saeed
MKMark Krakauer
Mark Krakauer
MSMegan R. Silas
Megan R. Silas
HSHassan Shah
Hassan Shah
To evaluate the performance of general-purpose, retrieval-augmented, and medicine-specific AI chatbots in answering common thyroid eye disease (TED) patient questions. Cross-sectional comparative evaluation. Online TED forum discussion posts were collected and synthesized into 15 representative patient questions across five groups spanning clinical (treatment, diagnosis, management, epidemiology) and non-clinical topics, grouped into three difficulty levels. Three differing large language models (LLMs) generated responses were randomized and anonymized for blinded assessment. Three oculoplastic surgeons evaluated clinical metrics; 3 medical students assessed non-clinical metrics. Three AI models generated responses: GPT-4o-mini (ChatGPT), a retrieval-augmented generation model grounded in TED literature (ChatGPT-RAG), and a specially trained LLM for healthcare professionals (OpenEvidence). Blinded raters assessed randomized responses. Statistical analysis used paired Wilcoxon signed-rank tests with Hedges' g for effect sizes. Clinical evaluation of responses was conducted using a 7-point Likert scale for relevance, accuracy, balance, and scope. Non-clinical metrics of empathy, understandability, and readability were also assessed using validated tools. OpenEvidence significantly outperformed both ChatGPT (mean clinical score 5.96 vs 4.94; Hedges' g = 1.21, P < 0.001) and ChatGPT-RAG (5.96 vs 5.55; g = 0.53, P < 0.001) in clinical rankings and across most clinical metrics, including accuracy and relevance. However, performance patterns reversed for non-clinical metrics, with ChatGPT consistently outperforming specialized models in empathy, understandability, and actionability (18.4 vs 14.96 for OpenEvidence; g = 1.25, P < 0.001). Across both domains, ChatGPT-RAG achieved intermediate performance, more closely trailing OpenEvidence clinically (g = 0.53) and ChatGPT with respect to non-clinical metrics (g = 0.44). Limitations include a modest sample of raters and synthesized questions from online forums, which may affect generalizability. Specialized medical AI models may have better clinical accuracy, while general-purpose models may outperform in patient communication and accessibility. The development of retrieval augmented generation-based approaches combining clinical precision with effective communication represents a promising direction for AI-powered patient education in TED and, potentially, other complex conditions.

2 October 2025 in Journal of Clinical Medicine
Verified
Journal Article
Open access
PDF file
ARAbinaya Ramakrishnan
Abinaya Ramakrishnan
DRDavid Reilly
David Reilly
JSJames Sayre
James Sayre
PAParsa Asachi
Parsa Asachi
KKKameel Khabaz
Kameel Khabaz
JMJustin McWilliams
Justin McWilliams
Purpose: Renal angiomyolipomas (AMLs) are benign renal neoplasms that may lead to spontaneous hemorrhage. Transcatheter arterial embolization (TAE) is a nephron-sparing treatment option, yet data on predictors of hemorrhage and re-intervention remain limited. This study evaluates clinical and radiologic outcomes of TAE and identifies predictors of hemorrhage and repeat embolization. Materials and Methods: A retrospective review of 66 patients (69 AMLs) undergoing TAE between 2010 and 2024 was conducted. Clinical, radiological, and procedural variables were analyzed. Tumor size, vascularity, and aneurysmal features were assessed pre- and post-embolization. Logistic regression models identified predictors of hemorrhage and repeat TAE. Results: Pre-treatment tumor diameter was the only significant predictor of hemorrhage (p = 0.011), with a threshold of 6.8 cm yielding 84.6% sensitivity and 71.3% specificity. All hemorrhagic tumors measured ≥4 cm. Post-embolization tumor volume predicted repeat TAE (p = 0.001), with a 248 mL cutoff. TAE significantly reduced tumor diameter (−33.5%) and volume (−60%) (p < 0.001). Radiologic success was achieved in 97% of cases, with a durable success rate of 84%. Clinical success was 94%, and complications occurred in 7.2% of patients, including two major events. Conclusions: TAE is a safe and effective treatment for renal AMLs. Tumor diameter >6.8 cm is a strong predictor of hemorrhage, while larger post-embolization volumes predict the need for re-intervention. These findings challenge the conventional 4 cm treatment threshold and support more individualized management strategies incorporating tumor morphology and response to embolization.

27 October 2025 in Journal of Vascular and Interventional Radiology
Verified
Journal Article
Open access
SASammy Allaw
Sammy Allaw
KKKameel Khabaz
Kameel Khabaz
QYQian Yu
Qian Yu
OAOsman Ahmed
Osman Ahmed
This study aims to synthesize evidence on trans-arterial embolization (TAE) for adhesive capsulitis and related shoulder tendinopathies to quantify pooled changes in pain, range of motion (ROM), American Shoulder and Elbow Surgeons (ASES) scores, and safety outcomes. A systematic review identified eligible cohort studies. PubMed and Scopus were searched through March 30, 2025. Standardized mean change using raw-score standardization (SMCR) was pooled with random-effects models (REML) for pain scores. ROM and ASES scores were pooled using raw mean changes. Heterogeneity (I2), influence analyses, and Egger's tests quantified robustness and publication bias. Twelve single-arm studies (329 shoulders; mean age 54 ± 10 y; 67 % female) met inclusion. Adhesive capsulitis accounted for 84% of cases. For pain, SMCR was -1.48 at 15-30 d, -2.51 at 31-90 d, and -2.59 at 91-180 d (p < 0.0001). For ROM, the pooled raw external rotation angles increased by +11.67°, + 25.78° and + 36.47° at the same intervals (p < 0.05). ASES scores rose by 22.94, 38.39 and 58.79 points at the three time windows (p < 0.05). Technical success was 100%. Of 94 recorded adverse events, 87 were mild; seven moderate and no severe events were reported per SIR guidelines. TAE yields large improvements in pain, shoulder mobility, and function with an favorable safety profile for refractory adhesive capsulitis and shoulder tendinopathies. Evidence is limited by uncontrolled designs, short follow-ups, and high heterogeneity, highlighting the need for randomized controlled trials.

27 September 2024 in Computers in Biology and Medicine
Verified
Journal Article
Closed access
KKKameel Khabaz
Kameel Khabaz
JKJunsung Kim
Junsung Kim
RMRoss Milner
Ross Milner
NNNhung Nguyen
Nhung Nguyen
LPLuka Pocivavsek
Luka Pocivavsek

1 January 2021 in Annals of Vascular Surgery
Verified
Journal Article
Closed access
SDSanjeev S. Dhara
Sanjeev S. Dhara
MHMichael Hermsen
Michael Hermsen
KKKameel Khabaz
Kameel Khabaz
EAErin Abbott
Erin Abbott
TBTrissa A. Babrowski
Trissa A. Babrowski
LPLuka Pocivavsek
Luka Pocivavsek