The development of laparoscopic artificial insemination (LAI) exemplifies the convergence of veterinary surgery and reproductive biology. In 1982, Killeen and Caffery first reported successful intrauterine insemination in ewes using laparoscopy, marking the inception of this technique (9). Early LAI systems were bulky and technically demanding, requiring specialized surgical teams, which limited their adoption. During the 1990s, the systematic review published by Gourley and Riese signaled the maturation of the technique and its progression into commercial use (7). Their work detailed standardized procedures and identified critical success factors, laying the foundation for subsequent refinements. With advances in optical imaging, materials science, and anesthesiology in the early 21st century, LAI equipment became more portable and durable, while the surgical procedure itself became safer and faster (14, 15). Modern LAI is now highly standardized. Sathe’s procedural review provided comprehensive best practices, covering equipment selection, animal preparation, surgical techniques, and postoperative care (8). These refinements reduced procedure time to 5–10 min, substantially improving operational efficiency and animal welfare (8, 16) (Table 1).

Successful LAI begins with meticulous planning and adherence to standardized protocols. The technical recommendations published by Gibbons et al. now serve as authoritative global guidelines for LAI practice (3). The key steps include:

The success of laparoscopic artificial insemination (LAI) is shaped by multiple interacting factors. Recent large-scale studies have provided scientific evidence to quantify their relative importance. Hill et al., analyzing 28,447 commercial LAI cases, reported an overall pregnancy rate of 71.7% and identified key determinants, including progesterone source, PMSG dosage, semen type (fresh vs. frozen–thawed), breed, and operator workload (10). More recently, Spanner et al. developed a multivariate prediction model based on 30,254 ewes, refining the analysis of predictors with modern statistical methods (11) (Table 2).

Estrus synchronization serves as the foundation for successful LAI, as its effectiveness directly determines insemination outcomes. Conventional protocols rely on combinations of progestogens and gonadotropins, yet new strategies continue to emerge as our understanding of small ruminant reproductive physiology deepens. Wildeus provided a seminal review outlining the theoretical framework of estrus synchronization in sheep and goats, emphasizing the pivotal role of PMSG in ovulation induction and the importance of tailoring dosages to breed, season, and individual variation (21). More recently, Habeeb highlighted the unique challenges of synchronizing seasonal breeders, further refining the theoretical basis for practical application (22). Field studies have confirmed these principles. Yu et al. compared five synchronization protocols in Hu sheep and demonstrated that Group II (11-day FGA + PGF2α on day 9 + 330 IU PMSG at withdrawal) achieved >80% estrus rate, the highest pregnancy outcomes, and up to 70% twinning rates, representing the most cost-effective solution for commercial farms (19). Tirpan et al. further showed that 300 IU PMSG was sufficient for synchronization, while higher doses increased the risk of excessive multiple births and dystocia (20). Earlier evidence by Mutiga et al. compared synchronization methods and PMSG dosages, providing historical support for dosage optimization (32). More recently, Shehabeldin et al. developed a PMSG-loaded chitosan nanoparticle system for controlled release, which enhanced synchronization efficiency and reduced the number of required injections, improving animal welfare (33). Alternative approaches have also been explored. Vallejo et al. compared Ovsynch with intravaginal progesterone devices in hair sheep and reported no significant differences in pregnancy rates, though Ovsynch was operationally simpler (34). Similarly, Deac et al. validated timed AI protocols during the non-breeding season, extending the applicability of LAI beyond seasonal constraints (35). In goats, Whitley and Jackson stressed that higher hormonal sensitivity requires species-specific dosage adjustments (36). A broader comparative review by Arya et al. summarized similarities and differences in synchronization strategies across cattle, sheep, and goats, offering insights for cross-species adaptation (37) (Table 3).

Traditional estrus detection relies heavily on visual observation, which is labor-intensive, subjective, and prone to errors. With advances in computer vision and artificial intelligence, automated estrus detection has emerged as a practical solution. Yu et al. developed a multi-target detection neural network based on an improved YOLO v3 architecture, incorporating K-means++ clustering and additional detection layers, achieving 98.56% precision and 98.04% recall in identifying ewe mounting behavior (38). This provides an accurate “time anchor” for LAI. Later, Yu et al. introduced a lightweight network (EfficientNet-B0 with Senet modules), delivering 99.44% accuracy with a model size of only 40.6 MB and 48.39 FPS, suitable for real-time deployment (39). Complementary innovations include non-contact biometric monitoring. Fuentes et al. integrated visible and infrared thermal imaging with machine learning algorithms to assess heat stress in sheep (40). Such tools not only aid estrus detection but also expand to animal health monitoring. Additionally, Hu et al. applied an enhanced YOLOv5 model to identify grazing behaviors, while Yang et al. reviewed applications of computer vision for body condition scoring (41, 42). Integrating multiple sensing modalities—such as ultrasonography, temperature loggers, and accelerometers—offers a pathway toward comprehensive estrus monitoring. These multimodal decision-support systems are shifting insemination decisions from experience-based to data-driven, enabling a transition from group-level scheduling to precise individual-level timing (39, 40).

Laparoscopic artificial insemination (LAI) and genomic selection [Genomic selection (GS)] form a complementary pair of technologies that collectively accelerate genetic progress in small ruminants. Genomic selection, which leverages dense SNP chip data to estimate genomic breeding values [Genomic estimated breeding values (GEBVs)], provides unprecedented accuracy in identifying elite sires. LAI then acts as a “genetic multiplier,” enabling these elite sires to disseminate their germplasm broadly across ewe populations, unrestricted by geography or time (43, 44). Van der Werf and Banks emphasized the transformative potential of GS in sheep breeding, particularly its ability to shorten generation intervals and improve selection accuracy. Similarly, Daetwyler et al. demonstrated the value of whole-genome sequencing for dissecting complex traits, while Moghaddar et al. confirmed that GS outperforms pedigree-based approaches in young animals (43, 44). The economic implications of GS integration with LAI are substantial. Shumbusho et al. evaluated genomic selection in meat sheep breeding programs, showing that GS combined with reproductive technologies generates considerable returns on investment (45). In dairy sheep, Baloche et al. validated the accuracy of genomic predictions in the French Lacaune breed, providing technical assurance for genetic progress in milk traits (46).

An optimal integration strategy involves:

Screening and selecting top young rams through GS.

This closed-loop “GS selection—LAI dissemination—data feedback” cycle maximizes genetic gain (45, 46).

Modern breeding programs must also align with sustainability goals. Brito et al. highlighted the importance of large-scale phenotyping to improve animal welfare and resilience (12). Likewise, González-Recio et al. incorporated methane emissions into dairy cattle breeding objectives, demonstrating the potential of genetic selection for environmental mitigation (13). Similar principles can be extended to small ruminants, where combining GS with LAI could yield breeds that are not only more productive but also climate-resilient and environmentally sustainable.

Artificial intelligence (AI) applications in LAI are transitioning from conceptual to practical. Spanner et al. developed a multivariate prediction model that integrates ewe characteristics, intraoperative uterine tone scores, and semen quality parameters to forecast pregnancy outcomes, achieving an AUC of 0.86 (11).

The model’s utility lies in several functions:

Individualized selection—identifying ewes with extremely low predicted fertility, thus avoiding unnecessary costs and animal stress.

Brito et al. emphasized that large-scale phenotyping is essential for training robust AI models with strong generalizability (12). More recently, Spanner et al. introduced a fertility model based on standardized in vitro semen thresholds, expanding AI-based predictive tools for field applications (29).

Ultimately, these AI systems are shifting LAI from an empirical practice to a data-driven, precision-guided intervention. The integration of genomic, environmental, and historical reproductive data will further enhance the personalization of reproductive management (12, 13) (Tables 4, 5).

Robotic-assisted surgery is already well established in human medicine and is gradually being adapted for veterinary use. These systems mitigate inherent limitations such as hand tremor and operator fatigue (47). For LAI, robotic systems offer several advantages:

Enhanced flexibility and precision through robotic arms that exceed the range of human motion.

Marescaux and Rubino described the ZEUS robotic platform in both experimental and clinical settings, highlighting its potential relevance for veterinary surgery (50). Although high costs remain a barrier, decreasing equipment prices and integration with AI and vision technologies could enable future “fully automated LAI workstations,” capable of completing insemination procedures from anesthesia to semen deposition with minimal human intervention.

Augmented reality (AR) overlays computer-generated information onto the real surgical field in real time, providing novel applications for LAI. Blixt and O’Brien systematically reviewed AR applications in veterinary surgery, reporting that AR can significantly enhance surgical accuracy and safety (51). In human medicine, Pratt et al. demonstrated that HoloLens-based AR systems can display three-dimensional (3D) vascular models intraoperatively, enabling surgeons to localize critical structures with greater precision (52). Similarly, Tepper et al. explored mixed reality integration, highlighting the combined potential of virtual and augmented reality in the operating room (53).

In LAI, AR applications could include:

Intraoperative navigation: Overlaying virtual markers to guide precise puncture sites within the uterine horns.

Advances in 3D imaging have provided a foundation for these AR applications. Fergo et al. demonstrated that 3D laparoscopy significantly improved depth perception and hand–eye coordination compared with 2D high-definition systems (54). Currò et al. further confirmed the value of 3D visualization for enhancing surgical precision (55). The combination of 3D/4 K laparoscopic imaging with AR guidance could transform LAI into a safer, more intuitive, and highly standardized procedure.

The Internet of Things (IoT) leverages distributed sensor networks to provide continuous monitoring of animals and their environment. Neethirajan reviewed wearable biosensors that capture physiological parameters such as body temperature, heart rate, activity levels, and rumination behavior (56). Wolfert et al. emphasized IoT as a key enabler of smart farming, where data-driven insights underpin precision livestock management (57).

In the context of LAI, IoT systems can:

Monitor ewe physiological indices (temperature, activity, feed intake) in real time using wearable devices.

Digital twin technology represents the most advanced form of IoT integration. Rasheed et al. described digital twins as dynamic, data-driven virtual replicas that evolve alongside their physical counterparts (58). Pylianidis et al. further highlighted their agricultural potential, noting their ability to integrate multi-source data into predictive models (59).

For LAI, digital twins could create individualized ewe models that incorporate genomic profiles, reproductive histories, real-time physiological data, and environmental conditions. These virtual models can:

Simulate different management strategies to predict reproductive outcomes.

This level of integration offers the potential for highly personalized reproductive management, transforming LAI from a population-level intervention into a truly individualized precision technology.

The economic evaluation of laparoscopic artificial insemination (LAI) requires a comprehensive assessment of direct costs, indirect benefits, and long-term impacts. Abbott conducted one of the earliest systematic cost–benefit studies on artificial insemination in wool sheep, concluding that LAI yields substantial returns on investment when the value of introduced germplasm is sufficiently high (60). Similarly, Valergakis et al. evaluated profitability in dairy sheep genetic improvement programs and found that LAI significantly enhanced milk yield and quality (61). More recent analyses by Myers et al. updated these models, incorporating the influence of modern technological advances on cost structures (6). Costs of LAI include equipment investment (laparoscopes, CO₂ insufflators, surgical tools), training and certification, semen procurement and storage, consumables, anesthesia and analgesics, and perioperative animal care (6, 60, 61). Benefits derive from higher pregnancy rates, improved semen utilization, rapid dissemination of elite genes, and reduced risks of sexually transmitted diseases compared with natural mating (6, 60, 61). Although single-procedure costs exceed those of transcervical insemination (TCAI), the superior pregnancy rates (60–70% vs. 20–40%) and long-term genetic and economic gains make LAI more profitable in structured breeding programs (6) (Table 6).

Despite its proven effectiveness, the widespread adoption of LAI is constrained by several challenges:

High technical threshold: LAI requires specialized surgical skills and extensive practice, making operator training a major bottleneck. Solutions include standardized certification systems, competency-based evaluation, and training with VR/AR simulation tools. Mentorship programs can accelerate skill transfer across generations (18, 62).

The industrialization of LAI depends on synergistic advances in research, policy, and market forces. The USDA Animal Genome Research Blueprint (2018–2027) emphasized the importance of genomics and reproductive technologies in advancing livestock health, productivity, and welfare (2). Similar frameworks highlight the role of policy in supporting LAI’s development through research funding, workforce training, and international collaboration. On the market side, consumer demand for high-quality animal products, environmental sustainability pressures, and global competition are driving adoption. As Brito et al. observed for dairy cattle, future livestock systems must balance productivity with animal welfare and ecological sustainability (64). These principles equally apply to small ruminants, underscoring LAI’s relevance as both productivity-enhancing and sustainability-enabling technology.

Future developments in LAI will be characterized by high levels of integration across hardware and software. At the hardware level, the next generation of intelligent LAI workstations will likely incorporate high-definition 3D laparoscopes, robotic arms, AI decision modules, and automated semen-handling systems. These platforms could deliver fully automated workflows, from animal preparation to postoperative monitoring, reducing reliance on specialist operators (50).

On the software side, integrated platforms will unify multi-source data—IoT sensor readings, genomic profiles, historical reproductive records, and environmental metrics—into centralized decision-support systems. By leveraging advanced AI algorithms, these platforms will provide individualized reproductive recommendations, effectively building a comprehensive digital reproductive twin for each ewe (56–58).

The adoption of precision medicine principles in veterinary science mirrors progress already achieved in human healthcare. In LAI, this will shift management from herd-level interventions to individualized protocols. Genomic data, obtained through whole-genome sequencing or SNP chips, will form the foundation of personalized reproductive planning by identifying genetic variants linked to fertility, disease resistance, or productivity traits (43, 44, 65–69). Epigenetic research will further expand this framework by uncovering how nutrition and environment regulate reproductive gene expression (70–76). Moreover, integrative multi-omics approaches—combining genomics, transcriptomics, proteomics, and metabolomics—will provide a systems-level view of reproductive physiology. These insights will enable targeted interventions that optimize insemination success while minimizing resource inputs (77–83).

The future of LAI must balance productivity gains with ecological sustainability. Thornton projected that livestock production will increasingly be judged by its environmental footprint (84), while Gerber et al. highlighted livestock’s central role in climate change mitigation (85).

LAI contributes to sustainability by:

Improving reproductive efficiency, thereby reducing the resource demand of maintaining breeding flocks.

Emerging methodologies such as life-cycle assessment [Life-cycle assessment (LCA)] will be essential for evaluating LAI’s environmental impact. Indicators like carbon footprint, water usage, and land-use efficiency will provide benchmarks to ensure that technological adoption aligns with sustainability goals (84, 85).