This study supported that the sheep cochlea was 2/3 the size of the human cochlea across all measurements, except for the internal auditory canal, where a ratio of 1/3 was observed. The number of spiral turns in the cochlea was identical in both species.

Prior to conducting research on humans, large animal studies are crucial, and this is especially true when investigating small anatomical systems. To the best of our knowledge, there are no previous studies that compare the volume of the IE between large animals and humans. High-resolution MRI proved to be a powerful modality for visualizing structures of the IE and comparing volume and dimensions between humans and sheep, enabling detailed quantitative analysis of these complex anatomical features. In the present study, the human cochlea had an average volume of 90.19 mm3, a length of 9.18 mm, and a width of 6.41 mm. These measurements correspond closely to those reported in the literature – 94.42 mm3 in volume [12], 9.2 mm in length, and 7 mm in width [13]. This study is the first to examine the sheep’s IE using MRI, revealing a size ratio of approximately 2/3 between sheep and humans. These findings are consistent with the previously described literature (Table 3).

For several years, otologists have shown interest in the sheep model, both for research and surgical training purposes [14]. While many research groups have focused on describing the anatomy of the sheep's middle ear [14,15,16,17,18,19,20,21,22], fewer have studied the IE anatomy [9, 14, 16, 17]. In the present study, the ratio of scala tympani length between sheep and humans was 0.68 (21.49 mm in sheep and 31.36 mm in humans). Histological studies in humans have measured a scala tympani length of 36 mm [6], which is greater than the value found in the present study. This discrepancy may be due to MRI resolution limitations, which make it challenging to visualize the cochlear apex. Beyond 29 mm in humans, the height of the scala tympani is less than 0.4 mm, matching the resolution threshold of MRI slices [6]. The resolution may also account for the slightly lower number of cochlear turns in humans (2.33 turns) in the present study, compared to the average 2.6 turns reported in the literature, with a range of 2.2–2.9 turns (described using method of casting temporal bone specimens) [23]. Additionally, differences in sheep strains may contribute to the variation seen across studies [21].

A significant difference was found in the IACL, with a ratio of 1/3 between sheep and humans. Only one previous study has evaluated the IACL in sheep using a CT-scan, reporting a mean of 2.00 mm and a ratio of 1/6 compared to humans [16]. This previous study identified an average human IACL of 13.0 mm, which closely aligns with the 12.41 mm observed in the present study [16]. Although young sheep were used in this study (mean age = 3.57 years, with a life expectancy of approximately 14 years [24]), sheep are fully mature by 24 months [24]. Therefore, it is unlikely that age influenced the smaller IACL measurements in sheep in the present study. This is supported by research showing that human IACL differs by only 0.2 mm between adolescence and adulthood [25].

Surgical dissections are essential for honing surgical skills because they help develop dexterity and improve outcomes in future procedures by minimizing errors. While cadaveric dissections are the gold standard for acquiring surgical proficiency, access to cadavers is limited in some countries due to high costs and strict regulations, creating challenges for training young otologists. In contrast, animal models offer a cost-effective, reliable alternative, with the ovine model proving particularly suitable. Anschuetz et al. developed an ex vivo endoscopic surgical training atlas using sheep [26]. Various procedures, including myringoplasty [27, 28], ossiculoplasty [26, 27], and stapedectomy [19, 29] can be practiced on this sheep model. Although sheep have a poorly pneumatized mastoid [15, 17, 19], the approach to the facial recess via mastoidectomy has been well documented [21], and cochlear implantation, using human electrode array, has been successfully performed on this model both ex vivo [9, 15, 30] and in vivo [31]. Overall, sheep represent a relevant surgical training model.

Given the anatomical and physiological similarities, sheep serve as an excellent model for auditory research. Some research groups are focusing on developing minimally invasive robotic cochlear implant devices to address hearing loss [32]. Others are exploring methods in sheep to restore hearing by improving the local delivery of therapeutic agents to the IE, such as hydrogels [33] or gas microbubble-assisted ultrasound [34]. While transgenic models in large animals are costly and more complex to develop than in small animals [11], transgenic sheep models are now available [35], although none have been specifically designed for IE pathologies. Beyond devices and procedures, the volume of IE fluids is a critical factor when comparing studies involving perilymph sampling for pharmacokinetic analysis, biomarker research, or drug delivery [11]. In this study, the contrast between the MRI signal intensity of the perilymph and endolymph was more pronounced in sheep, possibly indicating a difference in endolymphatic fluid composition between sheep and humans. Limited data exists in the literature comparing perilymph in humans and other species. Some studies have compared perilymph protein compositions, revealing a 59% similarity between humans and mice [36] and 64% between humans and guinea pigs [37]. Metabolomic studies have also been conducted [38], although these rely on human metabolomic databases [39], and no ovine-specific metabolomics database currently exists. Thus, extrapolating metabolomic findings from animal models to humans remains challenging.

Other large animal models can be considered for research purposes, but they present certain disadvantages. As for example, pigs have a thicker layer of soft and fatty tissues in the mastoid region, complicating the surgical approach for cochlear implantation [15]. Additionally, their weight, often around 300 kg, makes them more difficult to handle compared to sheep, which typically weigh around 70 kg [11]. Primates, such as macaques, offer a closer phylogenetic and anatomical resemblance to humans [40], but their use is highly restricted due to ethical regulations [30].

There were a few limitations to the present study. The same MRI parameters were used for both humans and sheep, with 0.4 mm thick slices. Given that the sheep's IE is approximately 2/3 the size of the human IE, these slices were insufficient to fully visualize the semicircular canals, making it impossible to measure the volume and length of the vestibule. To capture the semicircular canals in their entirety, 0.3 mm sections would be more suitable. This would also enable more accurate measurements of the cochlear turns and the length of the scala tympani. The development of 7 T MRI for humans and animals could significantly enhance the resolution of MRI acquisitions, thereby allowing for the distinct visualization of the various endocochlear compartments [41].