Artificial Intelligence in Healthcare Competition (TEKNOFEST-2021): Stroke Data Set.

OBJECTIVE: The artificial intelligence competition in healthcare was organized for the first time at the annual aviation, space, and technology festival (TEKNOFEST), Istanbul/Türkiye, in September 2021. In this article, the data set preparation and competition processes were explained in detail; the anonymized and annotated data set is also provided via official website for further research. MATERIALS AND METHODS: Data set recorded over the period covering 2019 and 2020 were centrally screened from the e-Pulse and Teleradiology System of the Republic of Türkiye, Ministry of Health using various codes and filtering criteria. The data set was anonymized. The data set was prepared, pooled, curated, and annotated by 7 radiologists. The training data set was shared with the teams via a dedicated file transfer protocol server, which could be accessed using private usernames and passwords given to the teams under a nondisclosure agreement signed by the representative of each team. RESULTS: The competition consisted of 2 stages. In the first stage, teams were given 192 digital imaging and communications in medicine images that belong to 1 of 3 possible categories namely, hemorrhage, ischemic, or non-stroke. Teams were asked to classify each image as either stroke present or absent. In the second stage of the competition, qualifying 36 teams were given 97 digital imaging and communications in medicine images that contained hemorrhage, ischemia, or both lesions. Among the employed methods, Unet and DeepLabv3 were the most frequently observed ones. CONCLUSION: Artificial intelligence competitions in healthcare offer good opportunities to collect data reflecting various cases and problems. Especially, annotated data set by domain experts is more valuable.

Manual stimulation of the whisker pad after hypoglossal-facial anastomosis (HFA) using a Y-tube conduit does not improve recovery of whisking function.

Facial nerve injury is a common clinical trauma involving long-term functional deficits with facial asymmetry leading to associated psychological issues and social hardship. We have recently shown that repair by hypoglossal-facial or facial-facial nerve surgical end-to-end anastomosis and suture [hypoglossal-facial anastomosis (HFA) or facial-facial anastomosis (FFA)] results in collateral axonal branching, polyinnervation of neuromuscular junctions (NMJs) and poor function. We have also shown that another HFA repair procedure using an isogenic Y-tube (HFA + Y-tube) and involving a 10-mm gap reduces collateral axonal branching, but fails to reduce polyinnervation. Furthermore, we have previously demonstrated that manual stimulation (MS) of facial muscles after FFA or HFA reduces polyinnervation of NMJs and improves functional recovery. Here, we examined whether HFA + Y-tube and MS of the vibrissal muscles reduce polyinnervation and restore function. Isogenic Y-tubes were created using abdominal aortas. The proximal hypoglossal nerve was inserted into the long arm and sutured to its wall. The distal zygomatic and buccal facial nerve branches were inserted into the two short arms and likewise sutured to their walls. Manual stimulation involved gentle stroking of the vibrissal muscles by hand mimicking normal whisker movement. We evaluated vibrissal motor performance using video-based motion analysis, degree of collateral axonal branching using double retrograde labeling and the quality of NMJ reinnervation in target musculature using immunohistochemistry. MS after HFA + Y-tube reduced neither collateral branching, nor NMJ polyinnervation. Accordingly, it did not improve recovery of function. We conclude that application of MS after hypoglossal-facial nerve repair using an isogenic Y-tube is contraindicated: it does not lead to functional recovery but, rather, worsens it.

Use of a Y-tube conduit after facial nerve injury reduces collateral axonal branching at the lesion site but neither reduces polyinnervation of motor endplates nor improves functional recovery.

BACKGROUND: Despite increased understanding of peripheral nerve regeneration, functional recovery after surgical repair remains disappointing. A major contributing factor is the extensive collateral branching at the lesion site, which leads to inaccurate axonal navigation and aberrant reinnervation of targets. OBJECTIVE: To determine whether the Y tube reconstruction improved axonal regrowth and whether this was associated with improved function. METHODS: We used a Y-tube conduit with the aim of improving navigation of regenerating axons after facial nerve transection in rats. RESULTS: Retrograde labeling from the zygomatic and buccal branches showed a halving in the number of double-labeled facial motor neurons (15% vs 8%; P < .05) after Y tube reconstruction compared with facial-facial anastomosis coaptation. However, in both surgical groups, the proportion of polyinnervated motor endplates was similar (≈ 30%; P > .05), and video-based motion analysis of whisking revealed similarly poor function. CONCLUSION: Although Y-tube reconstruction decreases axonal branching at the lesion site and improves axonal navigation compared with facial-facial anastomosis coaptation, it fails to promote monoinnervation of motor endplates and confers no functional benefit.

Hypoglossal-facial anastomosis (HFA) over a 10 mm gap bridged by a Y-tube-conduit enhances neurite regrowth and reduces collateral axonal branching at the lesion site.

PURPOSE: The outcome of severe peripheral nerve injuries requiring surgical repair (transection and suture) is usually poor. Recent work suggests that direct suture of nerves increases collagen production and provides unfavourable conditions for a proper axonal regrowth. We tested whether entubulation of the hypoglossal nerve into a Y-tube conduit connecting it with the zygomatic and buccal facial nerve branches would improve axonal pathfinding at the lesion site, quality of muscle reinnervation and recovery of vibrissal whisking. METHODS: For hypoglossal-facial anastomosis (HFA) over a Y-tube (HFA-Y-tube) the proximal stump of the hypoglossal nerve was entubulated and sutured into the long arm of a Y-tube (isogeneic abdominal aorta with its bifurcation). The zygomatic and buccal facial branches were entubulated and sutured to the short arms of the Y-tube. Restoration of vibrissal motor performance, degree of collateral axonal branching at the lesion site and quality of neuro-muscular junction (NMJ) reinnervation were compared to animals receiving HFA-Coaptation (no entubulation) after 4 months. RESULTS: HFA-Y-tube reduced collateral axonal branching. However it failed to reduce the proportion of polyinnervated NMJ and did not improve functional outcome when compared to HFA-Coaptation. CONCLUSION: Elimination of compression by tightly opposed nerve fragments improved axonal pathfinding. However, biometric analysis of vibrissae movements did not show positive effects suggesting that polyneuronal reinnervation - rather than collateral branching - may be the critical limiting factor. Since polyinnervation of muscle fibers is activity-dependent and can be manipulated, the present findings raise hopes that clinically feasible and effective therapies after HFA could be soon designed and tested.

The hypoglossal-facial nerve repair as a method to improve recovery of motor function after facial nerve injury.

Nerve repair after facial nerve injury provides neural input to the distal facial nerve and facial musculature via a variety of motor nerves such as hypoglossal, spinal accessory, masseteric branch of the trigeminal nerve and motor branches of the cervical plexus. The most commonly used procedure is the hypoglossal-facial nerve transfer. This cross-nerve paradigm is a unique nerve repair method as one motor nerve takes over the function of another motor nerve. The hypoglossal-facial nerve repair was reviewed by means of history, terminology, technical variations, and its capacity for recovery of function.

Method selection in craniofacial measurements: advantages and disadvantages of 3D digitization method.

AIM: Treatment of the craniofacial malformations is a primary goal of cranio-maxillo-facial surgeons. Surgical treatment of these malformations requires accurate data. Accuracy of measurement should be a priority of scientists to prevent statistical errors and therefore to promote the comparison of the results obtained from various research groups. In the present study, we aimed to compare three different measurement techniques, which were used frequently in craniofacial measurements. METHODS: A total number of 35 female and 35 male volunteer adults were included to the study. Two-dimensional (2D) photogrammetry, three-dimensional (3D) digitization and manual anthropometry methods were used for the present study. Measurements were obtained from the ear, eye, nose and face. RESULTS: By comparing three methods, our findings revealed that 3D digitization method is an easy, robust, and sensitive method to obtain the data. CONCLUSIONS: We think that 3D digitization method is accurate, and it can be applied to both clinical practice and research. Advantages and disadvantages of three methods are discussed with the relevant literature.

Morphological and ultrastructural analysis of the watershed zones after stripping of the vasa nervorum.

Peripheral nerve trunks are well-vascularized structures where a well-developed collateral system may compensate for local vascular damage. Vasculitis in nerve has a predilection for epineurial vessels and causes to the peripheral neuropathy, which is a major clinical feature of primary and secondary systemic vasculitides. In the present study, the goal was to simulate the vasculitic neuropathy in rat sciatic nerve and to investigate the watershed zones after stripping of the epineural vessels of the sciatic nerve. Sciatic function index values, light and electron microscopic evaluations of the experimental sciatic nerve suggested that the sciatic nerve was normal except for some watershed zones located in the peripheral part of the nerve. Although there is abundant collateral circulation in the peripheral nerve, distribution of the vessels of the watershed zones as observed in the present study should be elucidated by further studies.

Angiolipoma located inside the obturator canal and supplied by the umbilical artery.

During dissection of the retropubic region of a 55-year-old female cadaver, we encountered an angiolipoma located inside the obturator canal which was connected to the wall of the urinary bladder by a fibrous cord. The angiolipoma was supplied by a branch originating from the umbilical artery. Microscopically the benign soft tissue tumor was characterized by lobules of mature adipocytes and densely distributed networks of small and larger blood vessels, thus resembling typical histological features of an angiolipoma. Both the uncommon location of the angiolipoma and the abnormal branch of the umbilical artery entering the obturator canal should be taken into account during surgical procedures in this region, such as for orthopedic pelvic procedures, hernia repair or bladder/urethra-related interventions (e.g. transobturator tape, tension-free vaginal tape, colposuspension).