Fatma E. A. Hassanein

Prof. Dr. Fatma E. A. Hassanein

orcid

Research Focus & Metrics

Academic Career Length

3Years (2024-Present)

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

Representative Publications

View All Publications
30 October 2025 in Scientific Reports
Verified
Journal Article
Open access
PDF file
Fatma E. A. HassaneinFatma E. A. Hassanein
Fatma E. A. Hassanein
YAYousra Ahmed
Yousra Ahmed
SMShaymaa Maher
Shaymaa Maher
ABAhmed El Barbary
Ahmed El Barbary
AAAsmaa Abou-Bakr
Asmaa Abou-Bakr
Prompt design is a critical yet underexplored factor influencing the diagnostic performance of large language models (LLMs). Gemini Pro 2.5 shows promise in multimodal reasoning, but no prior study has systematically compared prompt structures in oral datasets against expert benchmarks. This study aimed to evaluate the diagnostic performance of a multimodal LLM (Gemini Pro 2.5) under different prompting strategies compared with oral medicine experts using prospective, histopathology-verified clinical vignettes. In a prospective, paired diagnostic accuracy study, Gemini pro 2.5 (a multimodal LLM) was evaluated under three prompting strategies: Direct (P-1), Chain-of-Thought (P-2), and Self-Reflection (P-3) on 300 oral lesion cases with histopathologic confirmation. Each prompt was applied to identical inputs and compared against diagnoses from board-certified oral medicine specialists. Accuracy, rubric-based narrative quality, probability calibration, and computational efficiency were assessed under STARD-AI guidelines. Human experts achieved the highest Top-1 accuracy (61%), but Chain-of-Thought prompting (P-2) led AI performance in Top-3 accuracy (82%) and produced the highest explanation quality (mean rubric score 8.49/10). No AI prompt matched human performance in low-difficulty cases. P-2 also showed the best calibration (Brier score 0.238) compared to P-1 and P-3. Resource-wise, Direct prompting was fastest, but longer outputs modestly improved Top-3 recall. Mixed-effects modeling confirmed that AI performance varied significantly by prompt structure, highlighting context-specific trade-offs. Prompt structure significantly affects the diagnostic performance and interpretability of AI-generated differentials in oral lesion diagnosis. While expert clinicians remain superior in straightforward cases, structured prompting, particularly Chain-of-Thought, may enhance AI reliability in complex diagnostic scenarios. These findings support the integration of prompt engineering into AI-assisted diagnostic tools to augment clinical decision-making in oral medicine.

1 July 2025 in Oral Diseases
Verified
Journal Article
Closed access
Fatma E. A. HassaneinFatma E. A. Hassanein
Fatma E. A. Hassanein
ABAhmed El Barbary
Ahmed El Barbary
RHRadwa R. Hussein
Radwa R. Hussein
YAYousra Ahmed
Yousra Ahmed
JEJylan El‐Guindy
Jylan El‐Guindy
AAAsmaa Abou‐Bakr
Asmaa Abou‐Bakr

24 February 2026 in Frontiers in Oral Health
Verified
Journal Article
Open access
PDF file
Fatma E. A. HassaneinFatma E. A. Hassanein
Fatma E. A. Hassanein
MAMalik Alkabazi
Malik Alkabazi
MTMelek Tassoker
Melek Tassoker
YAYousra Ahmed
Yousra Ahmed
SASuliman A Alsaeed
Suliman A Alsaeed
AAAsmaa Abou-Bakr
Asmaa Abou-Bakr
Background: Diagnosing oral lesions from benign conditions to oral cancer remains challenging due to overlapping visual features and reliance on histopathology. Large language models (LLMs) can integrate textual and visual cues, but their diagnostic accuracy and clinical utility in real decision-making contexts remain uncertain. To systematically evaluate the diagnostic performance, clinical usefulness, and limitations of LLMs in identifying oral lesions. Methods: PubMed, CINAHL, Embase, Web of Science, and Google Scholar were searched to 20 July 2025. Eligible studies applied LLMs (e.g., ChatGPT, Gemini, DeepSeek, Copilot, Claude) for diagnosis or differential diagnosis of oral lesions using text, images, or multimodal inputs. Outcomes included diagnostic accuracy, agreement metrics, and qualitative assessments of explanation quality and clinical applicability. Risk of bias was assessed using an adapted QUADAS-2. Narrative synthesis was performed due to heterogeneity. Results: Seventeen studies (>1,200 cases) were included. Diagnostic accuracy ranged from 25%–96%, varying by model version, input modality, and lesion complexity. Multimodal inputs consistently improved performance, with Cohen's κ up to 0.85–0.90. Advanced models (GPT-4o, DeepSeek-R1, o1-preview) outperformed earlier versions and approached expert performance in some tasks, although specialists generally retained superior Top-1 accuracy. Clinical utility was highest when LLMs were used to structure differential reasoning, highlight red-flag features, and support communication, but limited in tasks requiring fine morphological interpretation or severity grading. Overall risk of bias was low to moderate. Conclusions: LLMs demonstrate variable diagnostic performance and context-dependent supportive utility as adjunctive tools in oral lesion assessment, particularly in multimodal settings. They should complement, rather than replace, expert clinical judgment. Future research should prioritize real-world workflow evaluation, standardized prompting strategies, and prospective clinical validation. Systematic Review Registration: https://www.crd.york.ac.uk/PROSPERO/view/CRD420251090315 , identifier CRD420251090315.

2 July 2025 in BMC Oral Health
Verified
Journal Article
Open access
PDF file
AAAsmaa Abou-Bakr
Asmaa Abou-Bakr
EAEnji Ahmed
Enji Ahmed
HWHany William
Hany William
Fatma E. A. HassaneinFatma E. A. Hassanein
Fatma E. A. Hassanein
Background: Patients with head and neck cancer (HNC) receiving radiation therapy (RT) have complications affecting both general and dental health. While RT is effective against HNC, it negatively impacts oral tissues, including changes in periodontal attachment. This study aimed to evaluate the frequency of periodontal disease among HNC patients after RT in a sample of the Egyptian population, as well as to assess the associated risk factors. Materials and methods: A prospective cross-sectional study was conducted on 189 HNC patients attending a radiation center in Cairo, Egypt. Eligible patients underwent a full periodontal examination including plaque index (PI), bleeding on probing (BOP), clinical attachment level (CAL), and probing pocket depth (PPD). Subsequently, Clinical Oral Dryness Score (CODS), salivary flow rate and body mass index (BMI) were assessed to examine associations with periodontal health. Results: Periodontal disease was diagnosed in 183 (96.8%) HNC patients. Among them, 174 (95.08%) cases were diagnosed as periodontitis and 9 (4.9%) as gingivitis. The distribution of periodontitis different stages was as follows: Stage I was found in 21 (12.07%) cases, Stage II in 30 (17.24%), Stage III in 55 (31.61%), and Stage IV in 68 (39.08%). The correlation between the number of RT fractions and periodontal disease severity was not statistically significant (p = 0.837). However, there was a strong negative correlation between periodontitis stage and both body mass index (BMI) and salivary flow rate. In contrast, strong positive correlations were observed between periodontitis severity and RT dose, plaque percentage, and Clinical Oral Dryness Score (CODS). Conclusions: The frequency of periodontitis in the studied sample of HNC patients post RT was 95.08%, reflecting a notably high burden. The most prevalent periodontitis stage was the severe form (Stage IV). Higher periodontitis severity was found to be positively associated with RT dose, plaque percentage, and CODS, suggesting factors to plan future investigation on preventive care in HNC patients. While causality cannot be inferred due to the study design, these findings could be useful in developing more effective clinical management strategies in future research. Trial registration: The study was retrospectively registered on 29/10/2024 at ClinicalTrials.gov (NCT06667362).

6 February 2026 in BMC Oral Health
Verified
Journal Article
Open access
PDF file
Fatma E. A. HassaneinFatma E. A. Hassanein
Fatma E. A. Hassanein
RHRadwa R. Hussein
Radwa R. Hussein
YAYousra El sayed Ahmed
Yousra El sayed Ahmed
JEJylan El-Guindy
Jylan El-Guindy
DADoha E. Ahmed
Doha E. Ahmed
AAAsmaa Abou-Bakr
Asmaa Abou-Bakr
Background: The automated grading of clinical short-answer questions using large language models (LLMs) could alleviate faculty workload and improve the immediacy of feedback in dental education. However, evidence on the capacity of LLMs for rubric-based grading in dentistry remains limited. Therefore, this study aimed to compare the grading reliability and error patterns of two LLMs, ChatGPT-4 and the open-weight DeepSeek-3, against expert human evaluators. Materials and methods: In a retrospective cross-sectional study with comparative validation design, we analyzed 2,358 short-answer responses from 262 undergraduate dental students (across nine clinical questions). All responses were analyzed, then human-graded by three calibrated subject-matter experts (SME) (intraclass correlation coefficient [ICC] = 0.84) to provide a reference. Each LLM was provided a 12-point analytic rubric to guide the grading, but was not provided any prior examples of the grading task (i.e., a zero-shot prompt). We assessed agreement using ICC, Pearson correlation, Cohen’s kappa, and mixed-effects models, and examined error tiers (≤ 1, 2–3, > 3 points) across Bloom’s levels and response styles. Results: In this dataset, DeepSeek-3 obtained an ICC of 0.87 compared with ChatGPT-4 which obtained an ICC of 0.64. DeepSeek-3 matched exactly with human scores in 43.3% of cases and was within ± 1 point in 62.4%, compared with 35.5% and 44.1% for ChatGPT-4. High-error rates (> 3 points) were 7.5% for DeepSeek-3 vs. 26.9% for ChatGPT-4 (χ², p < 0.01). DeepSeek-3’s agreement was consistent across cognitive levels and response verbosity, while ChatGPT-4’s accuracy on higher-level and verbose responses was significantly lower ( p < 0.01). Both models exhibited an optimistic bias by over-scoring incorrect answers. Conclusions: DeepSeek-3 showed fewer large-magnitude errors and better agreement with human graders compared to ChatGPT-4, suggesting its potential value for large-scale AI-assisted assessment for dental education. Since both models can over-score on incorrect results, human-in-the-loop oversight is necessary for high-stakes applications. Further work should evaluate performance across more courses, institutions, and languages, as well as examine the effects of model calibration, bias reduction, and external validation before considering the broader integration of LLMs into dental education.

3 July 2025 in Journal of Cancer Research and Clinical Oncology
Verified
Journal Article
Open access
PDF file
Fatma E. A. HassaneinFatma E. A. Hassanein
Fatma E. A. Hassanein
CMChristine Mikhail
Christine Mikhail
SESarah Elkot
Sarah Elkot
AAAsmaa Abou-Bakr
Asmaa Abou-Bakr
Purpose Radiation-induced oral mucositis (RIOM) severely impacts patients with head and neck cancer (HNC) undergoing radiotherapy, often leading to pain and malnutrition. L-arginine and glutamine are immune-enhancing amino acids with potential benefits in wound healing and inflammation control. This study evaluated the efficacy of L-arginine versus L-glutamine oral suspensions in managing RIOM. Methods In this triple-blind, randomized controlled trial, 69 HNC patients with RIOM were allocated to three groups (n = 23 each): Group I (L-arginine 5 g + maltodextrin 5 g), Group II (glutamine 5 g + maltodextrin 5 g), or Group III (maltodextrin 10 g). Outcomes, assessed at weeks 2, 5, and 7 of radiotherapy, included the WHO oral mucositis scale, Pain Visual Analogue Scale (Pain-VAS), body mass index (BMI), and Oral Health Impact Profile (OHIP-14) questionnaire. Results By week 5, WHO scale scores differed significantly among groups (p < 0.001), with arginine and glutamine groups exhibiting lower mucositis severity than the maltodextrin group. Pain-VAS scores at weeks 5 and 7 were significantly lower in the arginine and glutamine groups compared to maltodextrin (p = 0.004 and p < 0.001, respectively). By 7th week of radiotherapy, BMI was significantly decreased in the maltodextrin group than in either the arginine (p = 0.028) or glutamine (p = 0.001) groups, indicative of treatment-mediated weight loss. In contrast, the BMI over time in the arginine (p = 0.87) and glutamine (p = 0.170) groups were almost constant. This indicates that compared to maltodextrin alone, both amino acid supplements prevented a decline in BMI during radiotherapy. OHIP-14 scores improved significantly in the arginine and glutamine groups at weeks 5 and 7 (p < 0.001), indicating better quality of life. Conclusions Both L-arginine and glutamine significantly reduced RIOM severity, pain, and weight loss compared to maltodextrin, while improving quality of life in patients with head and neck cancer. Although no statistically significant difference was found between the two, a higher proportion of patients receiving L-arginine achieved complete healing by week 7, suggesting a potential late advantage. These findings support the use of both amino acids as viable options for symptom management during radiotherapy. Trial registration ClinicalTrials.gov (NCT06764420), registered 08/01/2024.