Dorin Dumitru

Dorin Dumitru

Auto-generated profile, based on publicly available information

Dorin Dumitru

Breast Unit,Hull University Teaching Hospitals NHS Trust, Castle Road, Cottingham, HU16 5JQ, UK

Research Focus & Metrics

Academic Career Length

9Years (2018-Present)

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

Representative Publications

View All Publications
20 June 2025 in Journal of Medical Internet Research
Verified
Journal Article
Open access
ECEleanor Cheese
Eleanor Cheese
RBRaouef Ahmed Bichoo
Raouef Ahmed Bichoo
KGKartikae Grover
Kartikae Grover
DDDorin Dumitru
Dorin Dumitru
AZAlexandros Zenonos
Alexandros Zenonos
RPRebecca Pope
Rebecca Pope
Background: Having well-informed patients is crucial to enhancing patient satisfaction, quality of life, and health outcomes, which in turn optimizes health care use. Traditional methods of delivering information, such as booklets and leaflets, are often ineffective and can overwhelm patients. Educational videos represent a promising alternative; however, their production typically requires significant time and financial resources. Video production using generative artificial intelligence (AI) technology may provide a solution to this problem. Objective: This study aimed to use natural language processing (NLP) to understand free-text patient feedback on 1 of 7 AI-generated patient educational videos created in collaboration with Roche UK and the Hull University Teaching Hospitals NHS Trust breast cancer team, titled “Breast Cancer Follow Up Programme.” Methods: A survey was sent to 400 patients who had completed the breast cancer treatment pathway, and 98 (24.5%) free-text responses were received for the question “Any comments or suggestions to improve its [the video’s] contents?” We applied and evaluated different NLP machine learning techniques to draw insights from these unstructured data, namely sentiment analysis, topic modeling, summarization, and term frequency–inverse document frequency word clouds. Results: Sentiment analysis showed that 81% (79/98) of the responses were positive or neutral, while negative comments were predominantly related to the AI avatar. Topic modeling using BERTopic with k-means clustering was found to be the most effective model and identified 4 key topics: the breast cancer treatment pathway, video content, the digital avatar or narrator, and short responses with little or no content. The term frequency–inverse document frequency word clouds indicated positive sentiment about the treatment pathway (eg, “reassured” and “faultless”) and video content (eg, “informative” and “clear”), whereas the AI avatar was often described negatively (eg, “impersonal”). Summarization using the text-to-text transfer transformer model effectively created summaries of the responses by topic. Conclusions: This study demonstrates the success of NLP techniques in efficiently generating insights into patient feedback related to generative AI educational content. Combining NLP methods resulted in clear visuals and insights, enhancing the understanding of patient feedback. Analysis of free-text responses provided clinicians at Hull University Teaching Hospitals NHS Trust with deeper insights than those obtained from quantitative Likert scale responses alone. Importantly, the results validate the use of generative AI in creating patient educational videos, highlighting its potential to address the challenges of costly video production and the limitations of traditional, often overwhelming educational leaflets. Despite the positive overall feedback, negative comments focused on the technical aspects of the AI avatar, indicating areas for improvement. We advocate that patients who receive AI avatar explanations are counseled that this technology is intended to supplement, not replace, human health care interactions. Future investigations are needed to confirm the ongoing effectiveness of these educational tools.

1 June 2018 in Minerva Surgery
Verified
Journal Article
Closed access
DDDorin Dumitru
Dorin Dumitru
AKAyesha Khan
Ayesha Khan
GCGiuseppe Catanuto
Giuseppe Catanuto
NRNicola Rocco
Nicola Rocco
MNMaurizio B. Nava
Maurizio B. Nava
JBJohn R. Benson
John R. Benson
Article: Axillary surgery in breast cancer: the beginning of the end - Minerva Chirurgica 2018 June;73(3):314-21 - Minerva Medica - Journals

1 June 2021 in Future Oncology
Verified
Journal Article
Closed access
JBJohn R Benson
John R Benson
DDDorin Dumitru
Dorin Dumitru
IJIsmail Jatoi
Ismail Jatoi

15 February 2022 in Cancer Research
Verified
Journal Article
Open access
RBRaouef Ahmed Bichoo
Raouef Ahmed Bichoo
MEMohammed Bilal Elahi
Mohammed Bilal Elahi
DDDorin Dumitru
Dorin Dumitru
NINaila Bint Ihsan
Naila Bint Ihsan
EKEiman Khalifa
Eiman Khalifa
KGKritika Grover
Kritika Grover
Introduction: Sentinel lymph node biopsy represents the standard of care for staging axilla in breast cancer patients who are clinically node negative. ACOSOG Z11 trial results questioned the importance of low burden nodal disease. Results from the SOUND trial on omission of axillary surgery in selected patients with early stage breast cancer are still awaited. Breast screening programs aim to detect cancers at an early stage and are expected to have low burden of nodal disease. The aim of our study was to determine predictors of sentinel node biopsy positivity in invasive breast cancers detected on screening mammography. Methods: This retrospective study included patients with node negative invasive breast cancer detected on screening mammogram performed between April 2018 and March 2020 and subsequently underwent curative surgery including sentinel node biopsy at breast care unit of Castle Hill Hospital, Cottingham, U.K. We looked at patient’s clinical, radiological and pre-operative pathological features and correlated them with histological results of sentinel node biopsy. Statistical analysis of the data was performed using SPSS version 22 and p-value of 0.05 was considered significant. Results: We identified 304 patients who met inclusion criteria. The mean age of these patients was 62.71 ± 7.2 years (range 47-78 years). Majority (88%) of women were postmenopausal. Overall, the sentinel node positivity rate was 11.76%. Women over the age of 55 years were less likely to be node positive compared to younger women (9.9% vs 22%, p=0.016). Tumours less than 10mm in size had node positive rate of 5.5% compared to tumour larger than 21mm (25%, p = 0.000). In comparison to grade 2and grade 3 tumours, grade 1 tumours had lower node positivity rate, (5% vs 14.4%) but the difference did not reach statistical significance (p= 0.073). All women with tubular and mucinous histological subtypes (n=13) were node negative. All women with grade 1 tumours less than 10mm in size (n=48) had a negative sentinel node biopsy compared to women with grade 2 tumors more than 20mm (9/32=28% positivity rate) and p-value of 0.000. There was no association between oestrogen, progesterone and HER2 receptor status with sentinel lymph node positivity rate. Conclusion: In selected screen detected patients including those above the age of 55 years, having grade 1 cancer, and less than 10 mm size on pre-treatment imaging, there might be scope for omission of axillary surgery. Citation Format: Raouef Ahmed Bichoo, Mohammed Bilal Elahi, Dorin Dumitru, Naila Bint Ihsan, Eiman Khalifa, Masuma Sarker, Tapan K Mahapatra, Brendan Paul Wooler, Peter Kneeshaw, Kartikae Grover. Predictor of sentinel node positivity in screen-detected invasive breast cancer [abstract]. In: Proceedings of the 2021 San Antonio Breast Cancer Symposium; 2021 Dec 7-10; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2022;82(4 Suppl):Abstract nr P1-01-06.

1 June 2021 in Future Oncology
Verified
Journal Article
Closed access
JBJohn R Benson
John R Benson
DDDorin Dumitru
Dorin Dumitru
IJIsmail Jatoi
Ismail Jatoi

8 January 2025 in JMIR Preprints
Verified
Preprint
Open access
ECEleanor Cheese
Eleanor Cheese
RBRaouef Ahmed Bichoo
Raouef Ahmed Bichoo
KGKartikae Grover
Kartikae Grover
DDDorin Dumitru
Dorin Dumitru
AZAlexandros Zenonos
Alexandros Zenonos
RPRebecca Pope
Rebecca Pope
BACKGROUND Having well-informed patients is crucial to enhancing patient satisfaction, quality of life, and health outcomes, which in turn optimizes health care use. Traditional methods of delivering information, such as booklets and leaflets, are often ineffective and can overwhelm patients. Educational videos represent a promising alternative; however, their production typically requires significant time and financial resources. Video production using generative artificial intelligence (AI) technology may provide a solution to this problem. OBJECTIVE This study aimed to use natural language processing (NLP) to understand free-text patient feedback on 1 of 7 AI-generated patient educational videos created in collaboration with Roche UK and the Hull University Teaching Hospitals NHS Trust breast cancer team, titled “Breast Cancer Follow Up Programme.” METHODS A survey was sent to 400 patients who had completed the breast cancer treatment pathway, and 98 (24.5%) free-text responses were received for the question “Any comments or suggestions to improve its [the video’s] contents?” We applied and evaluated different NLP machine learning techniques to draw insights from these unstructured data, namely sentiment analysis, topic modeling, summarization, and term frequency–inverse document frequency word clouds. RESULTS Sentiment analysis showed that 81% (79/98) of the responses were positive or neutral, while negative comments were predominantly related to the AI avatar. Topic modeling using BERTopic with k-means clustering was found to be the most effective model and identified 4 key topics: the breast cancer treatment pathway, video content, the digital avatar or narrator, and short responses with little or no content. The term frequency–inverse document frequency word clouds indicated positive sentiment about the treatment pathway (eg, “reassured” and “faultless”) and video content (eg, “informative” and “clear”), whereas the AI avatar was often described negatively (eg, “impersonal”). Summarization using the text-to-text transfer transformer model effectively created summaries of the responses by topic. CONCLUSIONS This study demonstrates the success of NLP techniques in efficiently generating insights into patient feedback related to generative AI educational content. Combining NLP methods resulted in clear visuals and insights, enhancing the understanding of patient feedback. Analysis of free-text responses provided clinicians at Hull University Teaching Hospitals NHS Trust with deeper insights than those obtained from quantitative Likert scale responses alone. Importantly, the results validate the use of generative AI in creating patient educational videos, highlighting its potential to address the challenges of costly video production and the limitations of traditional, often overwhelming educational leaflets. Despite the positive overall feedback, negative comments focused on the technical aspects of the AI avatar, indicating areas for improvement. We advocate that patients who receive AI avatar explanations are counseled that this technology is intended to supplement, not replace, human health care interactions. Future investigations are needed to confirm the ongoing effectiveness of these educational tools.