Matan Friedman

Dr. Matan Friedman


Research Focus & Metrics

Academic Career Length

4Years (2023-Present)

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

Representative Publications

View All Publications
24 December 2025 in Journal of Clinical Medicine
Verified
Journal Article
Open access
PDF file
MFMatan Friedman
Matan Friedman
ASAmit Slouk
Amit Slouk
NGNoa Gonen
Noa Gonen
LGLaura Guzy
Laura Guzy
YPYael Ganor Paz
Yael Ganor Paz
IKIlia Kleiner
Ilia Kleiner
Background/Objectives: Large Language Models (LLMs), including ChatGPT-4 and Gemini, are increasingly incorporated into clinical care; however, their reliability within maternal–fetal medicine (MFM), a high-risk field in which diagnostic and management errors may affect both the pregnant patient and the fetus, remains uncertain. Evaluating the alignment of AI-generated case management recommendations with those of MFM specialists, emphasizing accuracy, agreement, and clinical relevancy. Study Design and Setting: Cross-sectional study with blinded online evaluation (November–December 2024); evaluators were blinded to responder identity (AI vs. human), and case order and response labels were randomized for each evaluator using a computer-generated sequence to reduce order and identification bias. Methods: Twenty hypothetical MFM cases were constructed, allowing standardized presentation of complex scenarios without patient-identifiable data and enabling consistent comparison of AI-generated and human specialist recommendations. Responses were generated by ChatGPT-4, Gemini, and three MFM specialists, then assessed by 22 blinded board-certified MFM evaluators using a 10-point Likert scale. Agreement was measured with Spearman’s rho (ρ) and Cohen’s (κ); accuracy differences were measured with Wilcoxon signed-rank tests. Results: ChatGPT-4 exhibited moderate alignment (mean 6.6 ± 2.95; ρ = 0.408; κ = 0.232, p < 0.001), performing well in routine, guideline-driven scenarios (e.g., term oligohydramnios, well-controlled gestational hypertension, GDMA1). Gemini scored 7.0 ± 2.64, demonstrating effectively no consistent inter-rater agreement (κ = −0.024, p = 0.352), indicating that although mean scores were slightly higher, evaluators varied widely in how they judged individual Gemini responses. No significant difference was found between ChatGPT-4 and clinicians in median accuracy scores (Wilcoxon p = 0.18), while Gemini showed significantly lower accuracy (p < 0.01). Model performance varied primarily by case complexity: agreement was higher in straightforward, guideline-based scenarios and more variable in complex cases, whereas no consistent pattern was observed by gestational age or specific clinical domain across the 20 cases. Conclusions: AI shows promise in routine MFM decision-making but remains constrained in complex cases, where models sometimes under-prioritize maternal–fetal risk trade-offs or incompletely address alternative management pathways, warranting cautious integration into clinical practice. Generalizability is limited by the small number of simulated cases and the use of hypothetical vignettes rather than real-world clinical encounters.

22 September 2023 in Journal of Clinical Medicine
Verified
Journal Article
Open access
PDF file
OGOhad Gluck
Ohad Gluck
EBElad Barber
Elad Barber
MFMatan Friedman
Matan Friedman
OFOhad Feldstein
Ohad Feldstein
OTOri Tal
Ori Tal
RSRon Sagiv
Ron Sagiv
Objective: To study whether the interval between gestational age calculated using the last menstrual period (GA-LMP) and gestational age calculated via ultrasound (GA-US) is correlated with the success rate of medical treatment in cases of miscarriages. Methods: This was a retrospective cohort study conducted in a gynecology unit in a tertiary medical center. Women who underwent medical treatment with Misoprostol for miscarriage at the Edith Wolfson Medical Center between 07/2015 and 12/2020 were included. Incomplete or septic miscarriages, multiple pregnancies, patients with irregular periods, and cases of missing data were excluded. Failure of medical treatment was defined as the need for surgical intervention due to a retained gestational sac, severe bleeding or retained products of conception. The cohort study was divided into two groups: patients with successful treatment and patients for whom surgical intervention was eventually needed. We performed both a univariate and multivariate analysis in order to identify whether a correlation between GA-LMP and GA-US interval is indeed a factor in the success rate of a medical abortion. Results: Overall, 778 patients were included in the study. From this cohort 582 (74.9%) had undergone a successful medical treatment, while 196 (25.1%) required surgical intervention due to the failure of medical treatment, as defined above. The GA-LMP to GA-US interval (in weeks) was 2.6 ± 1.4 in the success group, while the GA in the failure group was 3.1 ± 1.6 (p < 0.001). After performing a multivariant regression analysis, we were able to show that the GA-LMP to GA-US interval was found to be independently correlated with an increase in the treatment failure rate (aOR = 1.24, CI 95% (1.01–1.51), p = 0.03). Conclusions: In cases of miscarriage, longer GA-LMP to GA-US interval has been shown to be an independently correlated factor to lower success rate of the medical treatment option.

5 December 2024 in Obstetrics & Gynecology
Verified
Journal Article
Closed access
MFMatan Friedman
Matan Friedman
LMLiat Mor
Liat Mor
RSRotem Shazar
Rotem Shazar
NPNatalie Paul
Natalie Paul
RKRam Kerner
Ram Kerner
OGOhad Gluck
Ohad Gluck
OBJECTIVE: To compare the rates of treatment failure in cases of early pregnancy loss between mifepristone–misoprostol and misoprostol only. METHODS: This retrospective cohort study included patients who received medical treatment for early pregnancy loss between 2016 and 2023 at a single medical center. Patients returned for a follow-up ultrasonogram after 1 week and were treated again with misoprostol if needed. Finally, they were instructed to obtain an ultrasonogram after menstruation and to return for evaluation in case retained product of conception was suspected. We defined treatment failure as needing any surgical intervention because of retained product of conception, including cases when retained product of conception was diagnosed and treated after menstruation. In May 2022, we changed our protocol for treating early pregnancy loss from misoprostol to mifepristone and misoprostol. We compared the failure rate between patients who received mifepristone–misoprostol and those treated with misoprostol only. RESULTS: A total of 999 patients were included: 224 in the mifepristone–misoprostol group and 775 in the misoprostol-only group. The rate of treatment failure was significantly lower in the mifepristone–misoprostol group compared with the misoprostol-only group (17.8% vs 25.1%, P=.002). After multivariant analysis was performed, the use of mifepristone and misoprostol was associated with a reduction of 34% in the odds ratio for treatment failure compared with misoprostol alone (adjusted odds ratio 0.661, 95% CI, 0.44–0.97, P=.038). In addition, prior vaginal delivery was associated with a lower risk for treatment failure, and increasing gestational age (according to ultrasonogram) was correlated with a higher risk for treatment failure. CONCLUSION: The addition of mifepristone to misoprostol was associated with a significantly lower rate of treatment failure, including late surgical intervention for early pregnancy loss, compared with misoprostol alone.

15 March 2025 in American Journal of Obstetrics & Gynecology MFM
Verified
Journal Article
Closed access
MFMatan Friedman
Matan Friedman
LMLiat Mor
Liat Mor
ISIrit Segman
Irit Segman
YMYossi Mizrachi
Yossi Mizrachi
NSNoa Ben Shushan
Noa Ben Shushan
GBGiulia Barda
Giulia Barda

10 April 2023 in Journal of Clinical Medicine
Verified
Journal Article
Open access
PDF file
MZMasha Ben Zvi
Masha Ben Zvi
MCMaya Arad cohen
Maya Arad cohen
MFMatan Friedman
Matan Friedman
HHHadas Ganer Herman
Hadas Ganer Herman
EWEran Weiner
Eran Weiner
SGShimon Ginath
Shimon Ginath
Introduction and objective: This cross-sectional study aimed to compare the prevalence of urinary symptoms in physically active females to the general population represented by medical staff. Materials and methods: We conducted a survey utilizing the UDI-6 questionnaire on women playing catchball for at least one year and training twice a week or more in an official Israeli competitive league. The control group consisted of women practicing medicine (physicians and nurses). Results: The study group consisted of 317 catchball players and the control group consisted of 105 medical staff practitioners. Both groups were similar in most of the demographic characteristics. Urinary symptoms represented by UDI-6 scores were higher in women in the catchball group. Frequency and urgency symptoms were common in women playing catchball. Stress urinary incontinence (SUI) was insignificant between the groups (43.8% in the catchball group and 35.2% in the medical staff group, p = 0.114). However, severe symptoms of SUI were more common in catchball players. Conclusions: The rates of all urinary symptoms were higher in in catchball players. SUI symptoms were common in both groups. However, severe symptoms of SUI were more common in catchball players.

1 January 2025 in Reproduction and Fertility
Verified
Journal Article
Open access
MFMatan Friedman
Matan Friedman
NJNadine Jaffe
Nadine Jaffe
DTDaniel Tairy
Daniel Tairy
MTMaya Torem
Maya Torem
ARArieh Raziel
Arieh Raziel
YMYossi Mizrachi
Yossi Mizrachi
Graphical Planned oocyte cryopreservation (POC) has become widely available, allowing women to circumvent age-related fertility decline. The aim of our study was to examine whether women who have undergone POC were able to correctly predict the chance of having a child with their cryopreserved oocytes. We conducted a telephone survey with 260 women who underwent POC at our center between January 2017 and December 2023. Participants were asked to estimate their chance of having at least one live birth in case they would use their cryopreserved oocytes. For each participant, we also calculated the chance of achieving at least one live birth based on her age at the last oocyte retrieval and the number of cryopreserved oocytes, according to a model published by Goldman and colleagues in 2017. The median estimated probability of achieving a live birth was 50%, while the median calculated probability was 75% (P < 0.001). Only 28.1% of the participants accurately estimated their chances. In conclusion, a large percentage of women undergoing POC underestimate the probability of achieving a live birth if they use their cryopreserved oocytes. Improved counseling is essential to provide comprehensive information about the probability of live birth and prevent women from undergoing unnecessary treatments. Lay summary: More and more women choose to freeze eggs in order to circumvent age-related fertility decline. In this study, we asked women to estimate their chance of having a child if they used their frozen eggs in the future. For each patient, the chance was also calculated based on a model taking into account the number of eggs she had frozen and her age during egg freezing. Surprisingly, women estimated their chance of having a child as being lower than it actually is. Our results highlight the importance of providing patients with clear information about treatment success to prevent them from undergoing unnecessary additional treatments.