The insula has long been an overlooked cerebral structure but is recently gaining attention due to its underlying involvement in neurological and psychological disorders [9, 14, 16]. Literature on the anatomy of the insula is scarce, with only a few studies having analysed its morphology and morphometry using radiological modalities [1, 3, 11, 13].

In the current study, insulae predominantly resembled a trapezoidal shape in both right and left cerebral hemispheres (Table 1). A study by Cunha Cabral et al. [6] also found insular shape to vary between trapezoidal and triangular, however, their study reported primarily triangular shaped insulae. Studies by Afif et al. [1], Afif and Mertens [2], and Atlasi et al. [3] describe the insula as being trapezoidal in shape. While studies by Guenot et al. [10] and Mavridis et al. [11] describe the insula as having a triangular shape. The present study also noted that the insula was predominantly trapezoidal in both males and females. These findings suggest that insular shape is not dependant on sex or laterality (Table 1). The current study is the first to report on shape in terms of sex and laterality.

The CIS was “well seen” in the majority of left and right cerebral hemispheres (Table 1). These values were distinctly higher compared to those obtained by Naidich et al. [13]. The CIS was also mostly “well seen” in both males and females (Table 1). It may be postulated that the CIS is a reliable anatomical landmark on sagittal T1- and T2-weighted MRIs to identify major insular sulci and gyri. Additionally, the CIS was noted to lie in continuity with the central cerebral sulcus, as described in the literature [4].

The ASG and PSG were nearly always well displayed in the left and right cerebral hemispheres (Table 1). These results compared favourably with those of Naidich et al. [13] and Atlasi et al. [3]. The ASG and PSG were also predominantly “well seen” in males and females, corroborating the studies of Cunha-Cabral et al. [6] and Wysiadecki et al. [20] who reported the ASG and PSG to be “well developed” in their cadaveric studies. This suggests that MRI scans are a reliable modality in which to view the insular gyri.

Amongst the short gyri, the MSG presented less frequently and displayed more variable visibility than other gyri. This difference was significant for sex (p = 0.004) (Table 1), which may be due to female insulae being typically smaller than male insulae [11]. The MSG was “not seen” in only a few cerebral hemispheres, concurring with the findings of Naidich et al. [13], but in contrast to those of Atlasi et al. [3]. In this study, the MSG was “well seen” in 66% of left insulae and 64% of right insulae, the frequencies of which were lower than the report of Naidich et al. [13]. On the contrary, cadaveric studies by Cunha-Cabral et al. [6] and Wysiadecki et al. [20] recorded the MSG to be “well developed” only in 41.7% and 50% of insulae, respectively. In addition, the MSG was “not seen” in very few male and female insulae of the present study (Table 1).

The current study found that the ALG was “well seen” far more often than the PLG for both cerebral hemispheres (Table 1), thereby corroborating the findings of Naidich et al. [13]. However, this differs from Atlasi et al. [3] who found that both the ALG and PLG were predominantly “well seen” in both cerebral hemispheres. The ALG was also “well seen” in the majority of male and female insulae (Table 1). These findings were supported by Cunha-Cabral et al. [6] and Wysiadecki et al. [20]. On the other hand, the PLG was only “well seen” in 50% or less in both males and females. Cunha-Cabral et al. [6] reported that 86.4% of PLG were “well developed” which does not reflect the results of this study. Visibility of the CIS, ASG, PSG, ALG and PLG was independent of sex and laterality (Table 1). The insular gyri pattern identified in this study concurs that described in the literature [1,2,3, 13].

The mean widths of the ASG, MSG, PSG, ALG, and PLG were comparable in the left and right cerebral hemispheres, thus corroborating the cadaveric study by Cunha-Cabral et al. [6] (Table 2). The mean widths of all five gyri were larger in males than in females for both cerebral hemispheres (Table 2). These findings align with those of Atlasi et al. [3], who found that the mean antero-posterior distance of the insular base was larger in males than in females for both cerebral hemispheres. Mavridis et al. [11] noted a similar finding and stated that this was “more or less expected” as males tend to have larger and heavier brains.

Of all the insular gyri, the PLG was found to have the smallest mean widths in both cerebral hemispheres, as well as in both sexes (Table 2). In contrast, the ASG and PSG presented with similarly larger mean widths in both right and left cerebral hemispheres and in males and females alike (Table 2). The large variability in MSG and PLG visibility could be a result of their much shorter widths, relative to the other insular gyri. The largely consistent appearance of the ASG, PSG, and ALG may be attributed to their longer widths. This may be useful in a clinical setting, particularly for pre-planning of surgical procedures within the insular region. A study by Shura et al. [16] investigated the structural connectivity of the insula with regard to functionality and noted that the MSG contains a division that separates the anterior and posterior functional subdivisions. This may be a contributing factor to the variation in visibility and width of the MSG.

Statistically significant p-values were yielded for the MSG and PLG in the left cerebral hemispheres between males and females, owing to the distinctly larger morphometric parameters in males (Table 2).

The mean width of the AL was markedly larger than that of the PL (Table 2). Previous studies on non-human primates have shown that the anterior region of the insula (ASG and anterior ventral insular area) is primarily associated with the inferior frontal, orbitofrontal, and anterior temporal cortices and has connections with the amygdala, ventral striatum as well as the cingulate, entorhinal, and adjacent periamygdaloid temporal cortices, while the posterior region of the insula (PLG and posterior ventral insular area) is primarily associated with the temporal cortices and has connections to the dorsal striatum and the adjacent parietal and somatosensory cortices, the supplementary motor area, as well as the primary vestibular cortex [7, 16]. The anterior insula is thereby associated with olfactory, gustatory, viscero-autonomic, homeostatic, and limbic functions, whereas the posterior insula is linked to somatosensory, skeletomotor, auditory and vestibular function [7, 12, 15]. Moreover, Phan et al. [15] reported that the anterior lobule was stimulated by feelings of happiness, sadness, fear, anger, and disgust, while the posterior lobule was only stimulated by fear. The larger width of the AL in this study may be attributed to the numerous functional areas present within the anterior insula. The mean widths of both the AL and PL were also relatively comparable within their respective right and left cerebral hemispheres (Table 2). However, the AL and PL widths were larger in males than in females for both cerebral hemispheres (Table 2). These findings are similar to those found by Atlasi et al. [3]. Levels of statistical significance were recorded for the AL width in the right cerebral hemisphere of males and females, as well as for the PL width in both cerebral hemispheres of males and females. The right anterior insula is involved in interoceptive awareness of homeostatic emotions including thirst, fatigue, pain, and heart rate. Increased accuracy in this subjective perception of the inner body is correlated with a larger amount of grey matter in the right anterior insular region [12]. The significant difference between males and females regarding the width of the right AL may be attributed to the general increased volume of grey matter in the right anterior insula and the fact that males tend to have anatomically larger and heavier brains compared to females [11, 12]. The differences in AL and PL widths between males and females may be attributed to the difference in brain sizes as well as the functional associations of each insular region.

Intra-observer agreement was fair to almost perfect for assessments regarding insular morphology (Table 3). The intra-observer reliability coefficients indicated that the morphometric parameters were recorded in a reliable way (Table 4).

To the best of the authors’ knowledge, no previous study has analysed the widths of the insular gyri using MRI scans. Therefore, this study plays an integral role in contributing to the current body of anatomical knowledge on the insula by providing morphological and morphometric data from a radiological perspective. The findings may aid clinicians in the diagnosis and treatment of insula-related abnormalities.

A limitation of this study may include the potential presence of degree of error due to the curvature of the insular cortex as well as the nature of the MRI and DICOM viewer software. Future studies may opt to include scans where all gyri are visible. Measurements may be more accurate if multiple MRI slices are used to reconstruct the insula as it will allow for a more comprehensive analysis of insular structures. Additionally, the use of a common criterion for classification is recommended to ensure consistency. The accessory gyrus and Heschl gyrus were excluded from this study as their presence was inconsistent and often difficult to accurately visualize due to the curved nature of the insula. The transverse gyrus was also excluded as it is best visualized from an inferior view, while this study used sagittal MRI scans only. Future studies may consider utilizing axial imaging to analyse the transverse gyrus. Additionally, demographical information such as age and race was not available for analysis and thus is a limitation of the study. The handedness of individuals was not considered in this study as that information was not available. Future research may also investigate the association of handedness with insular anatomy.