Morphology and Hemispheric and Gender difference of the Anterior Ascending Ramus and the Horizontal Ascending Ramus of the Lateral Sulcus

Broca’s area is composed of the pars opercularis (PO) and the pars triangularis (PTR) of the inferior frontal gyrus(cid:0)which was separated by the anterior ascending ramus of lateral sulcus (aals) and the horizontal ascending ramus of the lateral sulcus (hals). The morphometry of these two sulci maybe have potential effects on the various functions of Broca’s area. To explore the morphology variation, hemispheric and gender difference of these two sulci, so as to enable better understanding of their patterns and the functions of the Broca’s area. BrainVISA was used to reconstruct and parameterize these two sulci with 3D MR images of 90 healthy right-handed subjects. The 3D anatomic morphology of these two sulci was presented using 4 sulcal parameters, including average depth (AD), average width (AW), top length (TL) and bottom length (BL). The aals could be identied in 98.89% of the hemispheres, while the hals could be identied in 98.33%. The morphological patterns of these two sulci were categorized into four main types. There were no statistically signicant interhemispheric and gender differences in the frequency of the morphological patterns. There was statistically signicant interhemispheric difference in BL of the aals and AW of the hals. Signicant gender differences were found in the bilateral AD of the aals and the BL and TL of the hals in right hemisphere. Our results will not only provide structural basis for the functional studies that related to Broca’s area, but also be helpful for the precise positioning of special structures in neurosurgery. brain structure The neuroimaging in brain


Introduction
The Broca's area lying in the third frontal convolution and above the lateral sulcus, is primarily comprised of the pars triangularis PTR and the pars opercularies PO . The Broca's area is known to be a core region for speech production. A wealth of neuroimaging analyses showed that activity in the pars opercularis and the pars triangularis increased during speech tasks (Heim et al. 2008;Amunts et al. 2004).
And impairment to the posterior ventrolateral region of the left hemisphere could lead to dysfunction of speech production (Broca 1861). Generally, the typical clinical symptoms of Broca's aphasia are decreased speech uency and mild-to-moderate impairment of speech comprehension (Pearce 2009). The area causing Broca's aphasia has been debated for more than a century. It is widely accepted that the damage limited to classical Broca's area probably involves parts of Broca's area and some other adjacent structures at present (Fridriksson et al. 2015).
The current research about Broca's area mainly focuses on the study of Occidental populations, but differences in genetics, culture and environmental exposures may give rise to differences in brain structure and function between the Oriental and Occidental populations (Lou et al. 2019;Han and Ma 2014). According to SBM analysis, compared to Westerners, the Chinese population had smaller structural measures in the frontal and parietal regions (Tang et al. 2018). In addition, the study of crosslanguage differences showed that the connections between the bilateral anterior temporal lobes and classical language areas are much stronger in a tonal (Chinese) language than in a nontonal (English) language, whereas the connections between the posterior temporal lobe and the inferior frontal lobe in left hemisphere is much stronger in a nontonal than in a tonal language during processing intelligible speech (Ge et al. 2015). It is obvious that cultural differences have been shown to affect linguistic functions of the brain (Xu et al. 2017). Therefore, it is necessary to utilize Chinese data to study Broca's area. Pars opercularis, namely area 44, is invoved in the advanced control of the orofacial musculature (Petrides 2016). Functional imaging studies of normal human subjects indicate that Brodmann area (BA) 44 is involved in syntax processing, whereas BA 45 mainly contributes to semantic processing (Goucha and Friederici 2015). Evidence indicated that BA 45 was active in controlled processes at the word-level such as semantic judgment or categorization, lexical-semantic access, and aspects of sentence-level semantic plausibility (Westphal et al. 2016;Petrides 2013). The functions of the Broca's area have promoted the anatomical studies of the asymmetry of this region. The anterior ascending ramus of the lateral sulcus and the anterior horizontal ramus of the lateral sulcus are the natural boundaries of Broca. The morphology of these two sulci are complex and their morphological changes perhaps cause changes in volume and cortical surface area of Broca. However, there are few studies on the anterior ascending ramus of the lateral sulcus (aals) and the horizontal ascending ramus of the lateral sulcus (hals), especially in Chinese.
Sex differences in the brain structure have been reported, ranging from global differences to local differences in regional tissue volume and the size of substructure (Luders et al. 2013;Xu et al. 2000). Kurth F (2017) found that the bilateral gray matter volume of BA 44 and BA 45 was signi cantly greater in females than in males, while the gender difference was not signi cant in BA 44/45 asymmetry (Kurth et al. 2017). Sulcal width and depth of men have been reported to change more rapidly, especially in the temproal collateral and cingulate sulci (Kochunov et al. 2005). However, there are few studies on gender differences of these two sulci.
In order to understand the in uence of these two sulci on the structure and function of Broca's area, in this study, we used in vivo brain MRI images with large sample size, advanced brain imaging and brain structure analysis methods. The present investigation was to characterize the morphology of these two sulci, to explore interhemispheric and gender difference on various morphometric parameters of these two sulci. It is expected to provide a method for complex brain sulcus analysis, provide anatomical basis for the function and related studies of Broca 's area, and provide basic guidance for functional neuroimaging studies and neurosurgical operations in related brain areas.

Materials And Methods
Subjects 90 Chinese volunteers (mean age 17.30±1.58 years, 50 males and 40 females) were recruited in the study. All subjects had no history of neurological and /or psychiatric illness, corresponding to Diagnostic and Statistical Manual of Mental Disorders (DSM-IV). All subjects were right-handed measured with the Edinburgh handedness inventory. The ethics approval was obtained from the ethics committee of Shandong University before the initiation of this study. All participants and their parents have given written informed consent.

MRI acquisition and processing
High quality three-dimensional structure MR data was acquired using a 3.0T GE (General Electric, Milwaukee, USA) MRI Scanner with a standard eight-channel head coil. Acquisition parameters for T1weighted structural MRI scan were identical and as follows: TR=18 ms; TE =10 ms; voxel size = 0.47 mm × 0.47 mm × 0.70 mm; FOV = 24.0 cm× 24.0 cm; matrix size =512 × 512; ip angle = 10°; slice thickness = 1.4 mm; slice gap = -0.7 mm; NEX = 2; total scan time = 12 min. All images were scanned along a horizontal line through the anterior and posterior commissures.

Image processing
First, all MR images were transformed to the standard space of the MNI305 atlas in FreeSurfer software (http://surfer.nmr.mgh.harvard.edu/fswiki/) using linear registration. Then the cortical sulci of subjects were reconstructed and automatically identi ed through BrainVISA BV software (http://brainvisa.info/). Particularly, cortical sulci were analyzed using the following steps: T1-weighted MRI import, bias correction, calculation of the mean whole brain mask, mesh the cerebral hemispheres, removal of nonbrain tissue and segmentation of brain tissues into white/gray matter(W/GM) and cerebrospinal uid (CSF), reconstruction of white/gray matter mesh , sulci recognition and automated labeling of sulci (Liu et al. 2010) (Fig .1). The labeling was rstly performed automatically then the result was checked visually as well as corrected manually if mislabeled sulci were found. Each step can be visually checked by an experienced neuroanatomist. For manually labeled MRI sections, all images were pre-processed by FreeSurfer rstly. The sections were taken at 1.4mm intervals. Using ITK-SNAP (http://www.itksnap.org/pmwiki/pmwiki.php), the manual annotation of the aals and hals was mainly completed in the sagittal section, while the results were corrected in the coronal section and transverse section. Moreover, the results of 3D reconstruction were compared with the result of BrainVISA to assist correction of the boundary of sulci.

Computation of sulci parameters
Measurements of sulcal length, sulcal width, and sulcal depth require the RIC tools for BrainVISA (http://ric.uthscsa.edu/personalpages/petr/). The average sulcal width (AW) for an individual sulcal structure was de ned as an average 3D distance between opposing gyral banks along the normal projections to the medial sulcal mesh (Kochunov et al. 2005) (Fig.2). The average sulcal depth (AD) was calculated as the average Euclidean distance between the points along the deepest edge of the sulcal skeleton surface and the nearest points on the outer edge of the sulcal ribbon. The sulcal length consist of top length (TL) and bottom length (BL). The top length was de ned as the length of the top edge of the sulcus. The bottom length was de ned as the length of the ridge at the base of the sulcus (Kochunov et al. 2008;Kochunov et al. 2005) (Fig.3).

Statistical analysis
The chi-squared test of independence was conducted to investigate the signi cance of interhemispheric and gender differences in the frequency with which the morphological patterns of Type I, Type II, Type III and Type . The statistical analysis of interhemispheric and gender differences of sulci parameter value were examined using one of two approaches. Independent-samples t-test was used to analyses sulci parameter values which satis ed the parametric test. Mann-Whitney test was used to examine sulci parameter values which satis ed the nonparametric tests. p < 0.05 was considered as statistically signi cant. The statistical analysis was conducted by the statistical software package SPSS (IBM SPSS, version 25).

Morphology of the aals and the hals
The aals could be observed in 98.89% of subjects (98.89% of left hemispheres and right hemispheres). And the hals could be observed in 98.33% of subjects (97.78% of left hemispheres and 98.89% of right hemispheres). The morphological patterns of these two sulci could be categorized into four main types ( Fig .4), the frequencies of which are presented in Table 1. Type could be identi ed in 55.56% of subjects (47.78% of left hemispheres and 63.33% of right hemispheres). The characteristics of this type was that the aals is clearly separated from the hals, forming the shape of the letter U ( Fig.4 a b). In 15(six left, nine right) of these hemispheres, showed that the aals and the hals formed the shape of the letter U on the lateral surface but formed the shape of the letter V on the depth of the hemisphere. In Type , observed in 13.33% of cases (16.67% of left hemispheres and 10% of right hemispheres), the aals and the hals sprang from a common point, forming the shape of the letter V ( Fig.4 c d). There are two cases in left hemisphere that the aals and the hals formed the shape of the letter V on the lateral surface but formed the shape of the letter U on the depth of the hemisphere. In Type , which was found in 12.78% of cases (13.33% of left hemispheres and 12.22% of right hemispheres), the aals shares a common stem with the hals, forming the shape of the letter Y ( Fig.4 e f). There was a third anterior ramus of the lateral ssure between the aals and the hals in Type ( Fig.4 g h). This type found in 7.78% of cases. Additional morphological patterns were identi ed with less frequency (Fig.5): 1) The lack of the aals was observed in two (2.22%) of the left and one (1.11 %) of the right hemispheres ( Fig.5 a b). 2) In 1.11% of cases (1.11% of left hemispheres and 1.11% of right hemispheres), the hals was absent (Fig 5. c d).
3) The aals was observed to bifurcate at its anterior end that was found in four (4.44%) of the left and one (1.11 %) of the right hemispheres ( Fig.5 e f) .4) The hals was observed to bifurcate at its anterior end that was found in six (6.67%) of the left and three (3.33 %) of the right hemispheres ( Fig.5 g h).
Interhemispheric Differences There was statistically signi cant difference (χ 2 =4.410, p<0.05) between the left and right hemispheres in Type I. However, this was no longer signi cant when a more conservative p value was used in a continuity correction, which is required for two-by-two contingency tables. In Type II, Type III and Type , no statistically signi cant difference was found between the left and right hemispheres for any of these comparisons. The results were shown in Table 1.
The results of interhemispheric differences about sulcal parameters were shown in Table 2,3. There was statistically signi cant interhemispheric difference in BL of the aals (U=2923.000, p<0.05).
Interhemispheric difference was statistically signi cant in AW of the hals (U=2818.000, p<0.05). There was no statistically signi cant interhemispheric difference in AW of the aals (U=2676.000, p=0.608).
Neither the aals nor the hals were statistically interhemispheric difference in TL (all p>0.05). The interhemispheric difference was no statistically signi cant in BL of the hals (U=3120.000, p=0.849). The AD of the aals and the hals were no statistically interhemispheric difference (all p>0.05).

Gender Differences
For each comparison, there was no statistically signi cant gender difference. The results were presented in Table 4. The results of gender differences about sulcal parameters were shown in The atlas which shown in supplementary materials presented a series of coronal, transverse, and sagittal sections from MRI transformed into the standard space of the MNI305. The four typical types of transverse section had following rules. The aals began to appear above the AC-PC level. Subsequently, the hals began to appear. The aals disappeared before the hals. When the aals and the hals appeared in the same section, the two branches were nearly horizontal parallel in Type I. In Type , the aals and the hals converged at an angle near the inner side of the brain, and the hals also gradually disappeared on the outer side of the brain after the aals disappeared. Similarly, the aals and the hals converged at an angle near the inner side of the brain in Type . However, as the sections deepen, the two sulci gathered at a sulcus. The third branch of the lateral sulcus occurred between the aals and the hals, and when the three sulci were in the same section, the three sulci were nearly parallel to each other in Type . The lateral side of the brain from anterior to posterior was the hals, the third branch of the lateral ssure, the aals.
When the two sulci were on the same plane, the two sulci also nearly horizontal paralleled. Four typical types can be distinguished by the sagittal plane, as shown in the supplementary material. The four typical types of consecutive coronal sections had following rules. When two branches appeared at the same level, they were nearly parallel in Type I. The hals got together with the aals in Type . In Type , the aals and the hals merged into a common stem. Generally, in Type , the hals and the third branch of the lateral ssure were nearly parallel in the same plane. When the hals disappeared, the aals appeared and was nearly parallel to the third branch of the lateral sulcus. The primary results of MRI sections were shown in Fig. 6-9. The consecutive transverse and coronal section of the atypical type had great variation and had not been analyzed in detail. The atypical type can be distinguished by the sagittal section.

Discussion
The present study determined the morphological variability of the anterior ascending ramus and the horizontal ascending ramus of the lateral sulcus. In our study, the morphological patterns of these two sulci were classi ed into typical and atypical types. The sulcal parameters were used to analyze hemispheric and gender differences of these two sulci.
The aals and the hals are two de ning sulci of the posterior ventrolateral frontal cortex of the human brain. These two sulci are well distinguished from the other adjacent sulcus considering the characteristics of the cortical surface and the depth of these two sulci.  Keller et al. 2007). The discrepancy maybe attributed to the great heterogeneity in the brain samples and the differences of sample sizes. The typical morphological patterns of these two sulci were classi ed into four types. Type was the typical morphology, the aals was clearly separated from the hals. A number of studies has been demonstrated (Keller et al. 2007).
Interestingly, in our study, some cases in Type showed that the aals and the hals formed the shape of the letter U on the lateral surface but formed the shape of the letter V on the depth of the hemisphere. The literature on this nding is still scarce. This may be because it is di cult to study these features in 3D data sets or because of the inconvenience of processing large amounts of data. The aals and the hals formed the shape of the letter V in Type . In Type , the aals shared a common trunk with the hals. The common trunk was either super cial or extended to the level of the insula. This founding was line with earlier studies which studied morphology and spatial probability maps of the horizontal ascending ramus of the lateral ssure. There were cases that a third anterior ramus of the lateral sulcus lied between the aals and the hals. This type observed in 16% of hemispheres in study of Sprung-Much(Sprung-Much and Petrides 2020). Unlike Sprung-Much's results, the present study only found 7.78% cases. We speculate that it may be ascribed to the potential differences between Chinese and Western brains. In the present investigation, the aals or the hals was observed to bifurcate at its anterior end. The results of Sprung-Much's study only showed that the hals was bifurcated at its anterior end. This may be caused by our inconsistent classi cation standards, or it may be due to our larger sample size. Overall, the morphology of these two sulci is complex and characterized by a high inter-subject variability. The results presented with analysis of the MRI Sections provided a detail way to distinguish the aals and the hals.
In our study, interhemispheric differences in the frequency of the morphological patterns of Type I, Type II, Type III, and Type were analyzed. As shown in Table 1, none of the comparisons were signi cant. The results were similar to those described by Sprung-Much et al (2020). Although there is no hemispheric difference in the morphological types of the two sulci, there are hemispheric differences in the measured values of these two sulci. The sulcal measured values can re ect the quantitative information of sulci. There was hemispheric difference in BL of the aals in our study, which may account for a higher g-SIs of the right hemisphere. Global sulcal indices (g-SIs) re ected the complexity of sulcal folds. Liu et al (2010), which used automated methods to explore global sulcal indices (g-SIs) of both cerebral hemispheres in elderly, found a higher g-SI of the right hemisphere than that of the left hemisphere (Liu et al. 2010). Evidence showed that the lengths of primary sulci could re ect the local pattern of the cerebral gyri cation (Imai et al. 2011), the hemispheric difference in BL of the aals may suggests hemispheric asymmetry of gyrus folding. Radiologists often use sulcal width as a measure of cerebral health in clinical practice (Hamelin et al. 2015). In the course of mild cognitive impairment and dementia, visual assessment of regional "widening" of cortical sulci (speci cally in the hippocampal, enthorhinal, and medial temporal cortices) are associated with neurocognitive decline (Bastos Leite et al. 2004). The interhemispheric difference in AW of the hals was signi cant, we raised the possibility that it might be potentially related to language lateralization. With regard to the relation between the interhemispheric difference of AW of the hals and language lateralization, it merits further investigation by functional imaging methods.
Our nding shows that sex difference of the four typical morphological patterns of the aals and the hals was no statistically signi cant difference. The literature on sex difference exploring morphological patterns of the aals and the hals has been scare, especially in Chinese population. This may be caused by limited research methods and insu cient sample size. The AD of the anterior ascending ramus of the lateral sulcus shows signi cantly gender difference in left hemisphere, as well as in right hemisphere. Sulcal depth of all sulci more strongly linked the volumes of adjacent gyri than global volumes. The difference of AD of the anterior ascending ramus of the lateral sulcus may be related to the asymmetry of the adjacent gyri, it can be indirectly inferred that there may be sex differences between gyri in Broca' s area. These results con rm women's advantages in relation to language, particularly in terms of uency and verbal memory tasks. The results support previous studies reporting sex differences in brain structure and function (Cosgrove et al. 2007;Frere et al. 2020;Wierenga et al. 2018).Our study found that female was characterized with longer TL and BL of the horizontal ascending ramus of the lateral sulcus in the right hemisphere, is in accordance with which found women had greater gyri cation than men of 20-30 years of age (Luders et al. 2004). The g-SI increase with an increase in number and/or area of sulcal folds. The TL and BL were closely related to the change of g-SI.

Limitation And Future Directions
Several potential limitations exist in our study. The sample size is a limitation of this study. The amount of data is large compared to other studies, but the analysis of neuroimaging often requires a larger amount of data. Another limitation of our study is that the results should be veri ed by the multiple methods. Although the accuracy of the results is analyzed by two methods of 3D reconstruction and section analysis, but we should use the multimodal method to better verify the results. We will increase the amount of data and use the multimodal method to perfect our results in next step.
In conclusion, the present study provides a detail description of the morphology of the anterior ascending ramus of the lateral sulcus and the horizontal ascending of the lateral sulcus. The morphology of these two sulci presented here can aid on improving sulcal identi cation, meantime, the present study reports a detailed analysis of difference of sulci quantitative information. The study sheds light for identifying these two sulci, it also plays a guiding role in the selection of surgical approach in the treatment of this area. Moreover, the results presented here provide anatomic basis for the exploration of functions of the Broca's area.

Declarations
Funding This work was supported by the National Natural Science Foundation of China (NO 81371533, 2017cxGC1501). Shuwei Liu approved to summit the manuscript.

Author contribution
Guarantors of integrity of entire study, Shuwei Liu, Yuchun Tang Ethical standards The authors declare that they have no competing nancial or non-nancial interests. All research was conducted in compliance with ethical standards.
Con icts of interest We declare that we have no con ict of interest.

Ethical approval
All procedures performed in studies involving human participants were in accordance with the ethical standards of the institutional .  Table 2 Inter-hemispheric differences of sulcal parameters of the aals.   Sulcal extraction and identi cation pipeline consists of (a) import the T1-weighted MRI into BrainVISA database; (b) bias correction; (c) calculate the mean whole brain mask; (d) meshed the cerebral hemispheres; (e) remove the non-brain tissue and segment the brain tissues into GM, WM and CSF; (f)reconstruction of white/gray matter mesh; (g) sulci recognition and (h) automated labeling of sulci.

Figure 2
The average sulcal width for an individual sulcal structure is measured as the average width of the intrasulcal space along the normal projections to the sulcal mesh.

Figure 3
The average sulcal depth was calculated as the average Euclidean distance between the points along the deepest edge of the sulcal skeleton surface and the nearest points on the outer edge of the sulcal ribbon.
The top length was de ned as the length of the top edge of the sulcus. The bottom length was de ned as the length of the ridge at the base of the sulcus.

Figure 4
Examples of morphological patterns formed by the anterior ascending ramus of the lateral sulcus (aals) and the horizontal ascending ramus of the lateral sulcus (hals). Aals, the red sulci; hals, the blue sulci; the third branch of the lateral sulcus, the pale-yellow sulci; inferior precentral sulcus (iprs), the green sulci, the sulcus diagonalis, the yellow sulci; the inferior frontal sulcus (ifs), the purple sulci. a,b :Type was that the aals is clearly separated from the hals, forming the shape of the letter U;c,d: Type was that the aals and the hals sprang from a common point, forming the shape of the letter V; e,f :Type was that the aals shares a common stem with hals , forming the shape of the letter Y;g,h: Type was a third anterior ramus of the lateral ssure between the aals and the hals .a,c,e,g: cortical map following sulci reconstruction; b,d,f,h: the pictures of Sulci detail.

Figure 5
Additional examples of morphological patterns formed by the anterior ascending ramus of the lateral sulcus (aals) and the horizontal ascending ramus of the lateral sulcus (hals). a,b, There is only one anterior ramus of the lateral sulcus; c,d, the horizontal ascending ramus of the lateral sulcus absent; e,f, the aals is bifurcated at its anterior end; g,h, the hals is bifurcated at its anterior end.

Figure 6
The left hemisphere is in MRI sections of Type . The rst line was the primary outcomes of coronal sections, the second line was the primary outcomes of horizontal sections, the third line was the primary outcomes of sagital sections. Red represents the aals, blue represents the hals.