Fig 1.
Simulating optimal irrigation in the Sylvian fissure.
We modeled the geometry of the lateral sulcus (A) using a plane 80×90 network connected to first neighbors (B). The nodes at the sulcus contour (white) were set as fluctuating sinks of identical mean value qfi ∼ N(1,σ); a source qin was set at the entrance of the sulcus (green) and sinks qfr and qte that represent the outflow towards the frontal and temporal lobes (yellow) were set to satisfy the conservation of flow, qin + qfr + qte + ∑qfi = 0. Two observables were computed to characterize the optimal transport networks (C): the distance d at which the MCA bifurcates and the number of loops.
Fig 2.
Dependence of network structure on vascular conditions.
We explored the effects of three variables in the structure of the network: (A) the fraction rout of the blood that leaves the Sylvian fissure, (B) the fraction rfr of the blood that leaves the fissure to irrigate the frontal lobe and (C) the current variability (standard deviation σ from unity currents feeding the fissure). For each ratio, we generated 200 locally optimal networks, for which we computed the bifurcation distance (the exit of the fissure is at distance 1) and the number of loops in the vascular network. Colors in the right bottom panel indicate independent loops.
Fig 3.
Anatomy of the insula and its limiting opercula.
The Sylvan fissure (arrows) is the most distinctive fold of the lateral hemispheric surface (A). Although it resembles other cortical sulci the Sylvian fissure runs deeper than any other and connects the lateral hemispheres with the basal cisterns. The frontal and temporal lobe regions limiting this fold are named opercula and bound the superior and inferior aspect of the fissure; its medial aspect is formed by the insula, a unique cortical region with distinctive convergent shallow folds (B). The insula and the opercula limit the divisions of the fissure into its principal components (C).
Fig 4.
Anatomy of the Middle Cerebral Artery (MCA).
The cortical branches of the middle cerebral arteries are depicted on this lateral view of a right hemisphere (A). As shown here, the MCA irrigates most of the lateral surface of the frontal and parietal lobe and the entire lateral temporal surface. Despite their proximity in the Euclidean space, no anastomotic vessels exist between the frontoparietal and temporal vessels and no arteries run across the Sylvian fissure. The distribution of the MCA vessels within the fissure are shown in (B). The lower trunk branches follow the superior aspect of the temporal lobe to reach the lateral subarachnoid space; the superior trunk climbs up the insular surface to make a complete turn at the top of the insular cleft. (C) and (D) show the apparent plexiform nature of the MCA branches that, on close examination, exhibit no anastomotic vessels or connections between the superior and the inferior trunk systems. (E) shows a lateral view of a right hemisphere in which the inferior frontal gyrus has been removed allowing a clear vision of the 180 degree turn of the MCA at the top of the insular cleft (arrows), the insular surface can be seen in the depth. On (F) the entire right hemisphere except for the central core and the inferior aspect of the temporal lobe has been removed. The 180˚ turn of the superior division of the MCA at the top of the insular cleft is again visible. The branches of the anterior cerebral artery are visible at the medial aspect of the left hemisphere.
Fig 5.
Development of the Sylvian fissure.
(A) represents an overview of the growth of the human brain in utero from post conceptional week 23 to 37. (B) represents the volumetric growth of the frontal lobe, central core, temporal and parietal lobes respectively. Logistic denotes the type of fit that was used to calculate the error in the figure (C) shows volumetric growth relative to total intracerebral volume at a given point. The insula and the central core grow at a slower rate than any other region and represent the smallest relative volume throughout gestation (ANCOVA controlling for side and gestational age; F(4, 173) = 123.61, p < 0.0001).
Fig 6.
The relative volume of the Sylvian fissure contracts over time.
(A) shows that the volume of the fissure (expressed as a proportion of the total intracranial volume) diminishes over time. A three-dimensional reconstruction and a coronal cut of the averaged MR images from the Gholipour atlas at weeks 24, 33 and 38 are depicted. The fissure volume is painted in light blue. (B) represents the absolute distance between the frontal and temporal lobes over time. The growth of these two lobes over the insula produces a graph in which the fronto-temporal distance diminishes over time, consistent with the relative volume contraction. These two lobes grow into each other and come in close contact only at the very end of gestation.