Development and technical validation of an artificial intelligence model for quantitative analysis of histopathologic features of eosinophilic esophagitis.

Background: In an attempt to provide quantitative, reproducible, and standardized analyses in cases of eosinophilic esophagitis (EoE), we have developed an artificial intelligence (AI) digital pathology model for the evaluation of histologic features in the EoE/esophageal eosinophilia spectrum. Here, we describe the development and technical validation of this novel AI tool. Methods: A total of 10 726 objects and 56.2 mm2 of semantic segmentation areas were annotated on whole-slide images, utilizing a cloud-based, deep learning artificial intelligence platform (Aiforia Technologies, Helsinki, Finland). Our training set consisted of 40 carefully selected digitized esophageal biopsy slides which contained the full spectrum of changes typically seen in the setting of esophageal eosinophilia, ranging from normal mucosa to severe abnormalities with regard to each specific features included in our model. A subset of cases was reserved as independent "test sets" in order to assess the validity of the AI model outside the training set. Five specialized experienced gastrointestinal pathologists scored each feature blindly and independently of each other and of AI model results. Results: The performance of the AI model for all cell type features was similar/non-inferior to that of our group of GI pathologists (F1-scores: 94.5-94.8 for AI vs human and 92.6-96.0 for human vs human). Segmentation area features were rated for accuracy using the following scale: 1. "perfect or nearly perfect" (95%-100%, no significant errors), 2. "very good" (80%-95%, only minor errors), 3. "good" (70%-80%, significant errors but still captures the feature well), 4. "insufficient" (less than 70%, significant errors compromising feature recognition). Rating scores for tissue (1.01), spongiosis (1.15), basal layer (1.05), surface layer (1.04), lamina propria (1.15), and collagen (1.11) were in the "very good" to "perfect or nearly perfect" range, while degranulation (2.23) was rated between "good" and "very good". Conclusion: Our newly developed AI-based tool showed an excellent performance (non-inferior to a group of experienced GI pathologists) for the recognition of various histologic features in the EoE/esophageal mucosal eosinophilia spectrum. This tool represents an important step in creating an accurate and reproducible method for semi-automated quantitative analysis to be used in the evaluation of esophageal biopsies in this clinical context.

Improved Postoperative Metrics with Modified Myofascial Closure in Fetal Myelomeningocele Repair.

BACKGROUND: The effect of modifications in fetal myelomeningocele (fMMC) closure techniques has not been extensively studied. OBJECTIVE: To study the effect of a modified closure technique on fMMC postnatal patient outcomes: hydrocephalus, hindbrain herniation, and cyst development. METHODS: We performed single-center retrospective study of a subset of post-MOMS (Management of Myelomeningocele Study) trial patients who underwent fMMC closure. After January 2015, the fetal myofascial closure technique was modified. Needlepoint monopolar cautery was used to raise dural lined myofascial flaps to create a more robust closure. Outcomes between the pre- and postmodification groups were compared with regard to hindbrain herniation, hydrocephalus, and cyst development. Families who transitioned care to local institutions were contacted via telephone for outcome information. RESULTS: From January 2011 to May 2016, data were reviewed from 119 fMMC closure patients. Patients without full follow-up data were excluded from the final analysis. Cerebrospinal fluid diversion was seen in 32 of 74 patients with the standard technique compared to 14 of 45 with the modified closure and was significantly decreased in postmodification when compared to that of the MOMS trial (P = .01). Hindbrain herniation resolution was significantly decreased in both the pre- and postmodification groups compared to that of the MOMS trial (P < .01). Prior to January 2015 with standard closure, 23 cysts required resection whereas no cysts required resection in the modified repair group (P < .01). CONCLUSION: Modified myofascial closure for fMMC closure is safe and feasible. The new approach reflects a decreased rate of cyst development requiring surgical resection, and a trend for improved rates of hindbrain herniation and hydrocephalus.

Detailed Analysis of Hydrocephalus and Hindbrain Herniation After Prenatal and Postnatal Myelomeningocele Closure: Report From a Single Institution.

BACKGROUND: The Management of Myelomeningocele Study (MOMS) demonstrated that fetal myelomeningocele (fMMC) closure results in improved hydrocephalus and hindbrain herniation when compared to postnatal closure. OBJECTIVE: To report on the outcomes of a single institution's experience in the post-MOMS era, with regard to hydrocephalus absence and hindbrain herniation resolution. METHODS: A single-center retrospective study of a subset of post-MOMS patients who underwent fetal/postnatal myelomeningocele closure was performed. Primary outcomes included cerebrospinal fluid (CSF) diversion status and hindbrain herniation resolution. Families were contacted via telephone for outcome information if care was transitioned to outside institutions. Univariate/multivariable analyses were performed using several prenatal and postnatal variables. RESULTS: From January 2011 to May 2016, data were reviewed from families of 62 postnatal and 119 fMMC closure patients. In the postnatal group, 80.6% required CSF diversion compared to 38.7% fetal cases (P < .01). Hindbrain herniation resolution occurred in 81.5% fetal repairs compared to 32.6% postnatal (P < .01). In the fetal group, fetal/premature neonatal demise occurred in 6/119 (5.0%) patients. There was a 42.0% decrease (95% CI -55.2 to -28.8) and 48.9% increase (95% CI 33.7 to 64.1) in risk difference for CSF diversion and hindbrain herniation resolution, respectively, in the fetal group. On univariate analysis for both groups, prenatal atrial diameter, frontal-occipital horn ratio, and hindbrain herniation resolution were significantly associated with the absence of clinical hydrocephalus. The treatment of hydrocephalus was significantly delayed in the fetal group compared to the postnatal group (10 mo vs 13.8 d). CONCLUSION: This study demonstrates the benefits of fMMC closure with regard to CSF dynamics.