Search

Mariko Jitsumori,1,* Shinya Matsuzaki,1,* Masayuki Endo,1 Takeya Hara,1 Takuji Tomimatsu,1 Satoko Matsuzaki,2 Tatsuya Miyake,1 Tsuyoshi Takiuchi,1 Aiko Kakigano,1 Kazuya Mimura,1 Eiji Kobayashi,1 Yutaka Ueda,1 Tadashi Kimura1 1Department of Obstetrics and Gynecology, Osaka University Graduate School of Medicine, Osaka, Japan; 2Department of Obstetrics and Gynecology, Otemae Hospital, Osaka, Japan*These authors contributed equally to this workCorrespondence: Shinya Matsuzaki; Aiko KakiganoDepartment of Obstetrics and Gynecology, Osaka University Graduate School of Medicine, 2-2 Yamadaoka, Suita, Osaka 565-0871, JapanTel +81-6-6879-3355Fax +81-6-6879-3359Email zacky@gyne.med.osaka-u.ac.jp; ai-kaki@gyne.med.osaka-u.ac.jpBackground: Uterine artery embolization (UAE) is used to treat severe postpartum hemorrhage (PPH). According to a few studies, UAE for PPH was associated with preterm birth, fetal growth restriction (FGR), and placenta accreta spectrum (PAS) in subsequent pregnancies. These previous studies, however, lacked controls, and to the best of our knowledge, no systematic literature reviews have been conducted thus far. We report the results of our retrospective case–control study of pregnancies after UAE at a single center and include a literature review to evaluate the risk of PAS in pregnancies after UAE.Methods: We retrospectively reviewed data from deliveries at our hospital between January 2012 and October 2017. We divided the delivery data into cases with previous UAEs performed for PPH (the post-UAE group) and those without UAEs (the non-UAE group, which included women without previous PPH). We defined PAS as cases in which hysterectomy was performed and pathological examination confirmed the diagnosis. Results are presented as odds ratios (ORs) with 95% confidence intervals (95% CIs).Results: We used data from 3155 patients in this study, of whom 16 patients had undergone UAE (post-UAE group) and 3139 had not (non-UAE group). We found no differences between the groups in terms of frequency of preterm births (12.5% versus 14.2%, respectively; OR, 0.863; 95% CI, 0.218 to 3.414; P = 0.84) or FGR (6.2% versus 10.0%, respectively; OR, 0.602; 95% CI, 0.104 to 3.584; P = 0.61). However, cases of PAS were significantly more common in the post-UAE group (37.5%) than in the non-UAE group (1.2%; OR, 50.303; 95% CI, 17.38 to 145.592; P < 0.01).Conclusion: Our results suggest that previous UAE is a significant risk factor for PAS.Keywords: placenta accreta, placenta accreta spectrum, uterine artery embolization, postpartum hemorrhage, obstetric outcomes

Microfossils of fish teeth and denticles, referred to as ichthyoliths, provide critical information for depositional ages, paleo‐environments, and marine ecosystems, especially in pelagic realms. However, owing to their small size and rarity, it is time‐consuming and difficult to analyze large numbers of ichthyoliths from sediment samples, limiting their use in scientific studies. Here, we propose a method to automatically detect ichthyoliths from microscopic images using a deep learning technique. We applied YOLO‐v7, one of the latest object detection architectures, and trained several models under different conditions. The model trained under appropriate conditions with an original data set achieved an F1 score of 0.87. We then enhanced the data set efficiently using the pre‐trained model. We validated the practical applicability of the model by comparing the number of ichthyoliths detected by the model with those counted manually. This revealed that the best model can predict the number of triangular teeth, denticles and irregularly shaped teeth with minimal human intervention. This object detection method can extend the applicability of deep learning to a wider array of microfossils and has the potential to dramatically increase the spatiotemporal resolution of ichthyolith records for applications across disciplines. Plain Language Summary: Fossils of fish teeth and denticles, referred to as ichthyoliths, can be used to study the environmental changes of marine conditions throughout Earth's history. However, it is time‐consuming and difficult to analyze large numbers of ichthyoliths from sediment samples, limiting their use in scientific studies. Here, we trained several artificial intelligence models to automatically detect ichthyoliths from microscopic images. The best model is suitable for counting the number of fish teeth, denticles, and irregularly shaped teeth fragments with minimal human intervention. We propose that object detection, a deep learning technique used in this study, can be applicable for the study of various microfossils, as well as for increasing the spatiotemporal resolution of ichthyolith records. Key Points: We trained object detection models under different conditions to detect microfossil fish teeth and denticles from microscopic imagesThe best model can count teeth, denticles and irregularly shaped teeth from samplesObject detection may improve the observation efficiency of a wide array of microfossils [ABSTRACT FROM AUTHOR]