Relative Afferent Pupillary Defect Screening Through Transfer Learning.

Abnormalities in pupillary light reflex can indicate optic nerve disorders that may lead to permanent visual loss if not diagnosed in an early stage. In this study, we focus on relative afferent pupillary defect (RAPD), which is based on the difference between the reactions of the eyes when they are exposed to light stimuli. Incumbent RAPD assessment methods are based on subjective practices that can lead to unreliable measurements. To eliminate subjectivity and obtain reliable measurements, we introduced an automated framework to detect RAPD. For validation, we conducted a clinical study with lab-on-a-headset, which can perform automated light reflex test. In addition to benchmarking handcrafted algorithms, we proposed a transfer learning-based approach that transformed a deep learning-based generic object recognition algorithm into a pupil detector. Based on the conducted experiments, proposed algorithm RAPDNet can achieve a sensitivity and a specificity of 90.6% over 64 test cases in a balanced set, which corresponds to an AUC of 0.929 in ROC analysis. According to our benchmark with three handcrafted algorithms and nine performance metrics, RAPDNet outperforms all other algorithms in every performance category.

Low Glasgow Coma Score in Traumatic Intracranial Hemorrhage Predicts Development of Cerebral Vasospasm.

BACKGROUND: The exact mechanism, incidence, and risk factors for cerebral vasospasm after traumatic intracranial hemorrhage (ICH) continue to be poorly characterized. The incidence of post-traumatic vasospasm (PTV) varies depending on the detection modality. OBJECTIVE: We aimed to shed light on the predictors, associations, and true incidence of cerebral vasospasm after traumatic ICH using digital subtraction angiography (DSA) as the gold standard. METHODS: We examined a prospectively maintained database of traumatic brain injury (TBI) patients to identify patients with ICH secondary to TBI enrolled between 2002 and 2015 at our trauma center. Patients with TBI-associated ICH and evidence of elevated velocities on transcranial Doppler and computed tomography angiograms, confirmed with DSA were included. The diagnostic cerebral angiograms were evaluated by 2 blinded neurointerventionalists for cerebral vasospasm. Statistical analyses were conducted to determine predictors of PTV. RESULTS: Twenty patients with ICH secondary to TBI and evidence of vasospasm underwent DSAs. Seven patients (7/20; 35%) with traumatic ICH developed cerebral vasospasm and of those, 1 developed delayed cerebral ischemia (1/7; 14%). Of these 7 patients, 6 presented with subarachnoid hemorrhage (6/7; 85%). Vasospasm was substantially more common in patients with a Glasgow Coma Scale <9 (P = 0.017) than in all other groups. CONCLUSIONS: PTV as demonstrated by DCA may be more common than previously reported. Patients who exhibit PTV were more likely to have a Glasgow Coma Scale <9. This subgroup of patients may benefit from more systematic screening for the development of PTV, and earlier monitoring for signs of delayed cerebral ischemia.

Internal Maxillary Artery Preoperative Embolization Using n-Butyl Cyanoacrylate and Pushable Coils for Temporomandibular Joint Ankylosis Surgery.

BACKGROUND: Temporomandibular joint (TMJ) ankylosis causes disability through impaired digestion, mastication, speech, and appearance. Surgical treatment increases range of motion with resultant functional improvement. However, substantial perioperative blood loss can occur (up to 3 L) if the internal maxillary artery (IMAX) is injured as it traverses the ankylotic mass. Achieving hemostasis is difficult because of limited proximal IMAX access and poor visualization. Our aim is to investigate the technical feasibility and preliminary safety of preoperative IMAX embolization in patients undergoing TMJ ankylosis surgery. METHODS: Case series using chart reviews of 2 patients who underwent preoperative embolization before TMJ ankylosis surgery. RESULTS: Both patients were women (28 and 51 years old) who had severely restricted mouth opening. Embolization was performed using general anesthesia with nasal intubation on the same day of TMJ surgery. Both patients underwent bilateral IMAX embolization using pushable coils (Vortex, Boston Scientific) of distal IMAX followed by n-butyl-cyanoacrylate (Trufill, Cordis) embolization from coil mass up to proximal IMAX. There were no complications from the embolization procedures. Both patients had normal neurologic examination results. TMJ surgery occurred with minimal operative blood loss (≤300 mL for each surgery). Maximum postoperative mouth opening was 35 mm and 34 mm, respectively. One patient had a postoperative TMJ wound infection that was managed with antibiotics. CONCLUSIONS: Preoperative IMAX embolization before TMJ ankylosis surgery is technically feasible with encouraging preliminary safety. There were no complications from the embolization procedures and surgeries occurred with low volumes of blood loss.