LERCause: Deep learning approaches for causal sentence identification from nuclear safety reports.

Identifying causal sentences from nuclear incident reports is essential for advancing nuclear safety research and applications. Nonetheless, accurately locating and labeling causal sentences in text data is challenging, and might benefit from the usage of automated techniques. In this paper, we introduce LERCause, a labeled dataset combined with labeling methods meant to serve as a foundation for the classification of causal sentences in the domain of nuclear safety. We used three BERT models (BERT, BioBERT, and SciBERT) to 10,608 annotated sentences from the Licensee Event Report (LER) corpus for predicting sentence labels (Causal vs. non-Causal). We also used a keyword-based heuristic strategy, three standard machine learning methods (Logistic Regression, Gradient Boosting, and Support Vector Machine), and a deep learning approach (Convolutional Neural Network; CNN) for comparison. We found that the BERT-centric models outperformed all other tested models in terms of all evaluation metrics (accuracy, precision, recall, and F1 score). BioBERT resulted in the highest overall F1 score of 94.49% from the ten-fold cross-validation. Our dataset and coding framework can provide a robust baseline for assessing and comparing new causal sentences extraction techniques. As far as we know, our research breaks new ground by leveraging BERT-centric models for causal sentence classification in the nuclear safety domain and by openly distributing labeled data and code to enable reproducibility in subsequent research.

Bat4RCT: A suite of benchmark data and baseline methods for text classification of randomized controlled trials.

Randomized controlled trials (RCTs) play a major role in aiding biomedical research and practices. To inform this research, the demand for highly accurate retrieval of scientific articles on RCT research has grown in recent decades. However, correctly identifying all published RCTs in a given domain is a non-trivial task, which has motivated computer scientists to develop methods for identifying papers involving RCTs. Although existing studies have provided invaluable insights into how RCT tags can be predicted for biomedicine research articles, they used datasets from different sources in varying sizes and timeframes and their models and findings cannot be compared across studies. In addition, as datasets and code are rarely shared, researchers who conduct RCT classification have to write code from scratch, reinventing the wheel. In this paper, we present Bat4RCT, a suite of data and an integrated method to serve as a strong baseline for RCT classification, which includes the use of BERT-based models in comparison with conventional machine learning techniques. To validate our approach, all models are applied on 500,000 paper records in MEDLINE. The BERT-based models showed consistently higher recall scores than conventional machine learning and CNN models while producing slightly better or similar precision scores. The best performance was achieved by the BioBERT model when trained on both title and abstract texts, with the F1 score of 90.85%. This infrastructure of dataset and code will provide a competitive baseline for the evaluation and comparison of new methods and the convenience of future benchmarking. To our best knowledge, our study is the first work to apply BERT-based language modeling techniques to RCT classification tasks and to share dataset and code in order to promote reproducibility and improvement in text classification in biomedicine research.

Autoimmune retinopathy with associated anti-retinal antibodies as a potential immune-related adverse event associated with immunotherapy in patients with advanced cutaneous melanoma: case series and systematic review.

OBJECTIVE: To demonstrate the spectrum of autoimmune retinopathy (AIR) associated with immunotherapy for advanced cutaneous melanoma. METHODS AND ANALYSIS: Retrospective chart review on patients with advanced cutaneous melanoma who developed AIR after initiating immunotherapy. Complete ophthalmic examination and relevant ancillary testing were performed on each patient. The presence of AIR-associated anti-retinal antibodies was confirmed by western blot and/or immunohistochemical staining. Ophthalmic and systemic outcomes after treatment for AIR were followed over time. A systematic review of AIR associated with immunotherapy for cutaneous or non-ocular mucosal melanoma was carried out in accordance with Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines. RESULTS: Case 1 developed photopsia and nyctalopia with electroretinographic findings characteristic for melanoma-associated retinopathy 1 week after initiating ipilimumab/nivolumab immunotherapy. Case 2 experienced new severe bilateral visual field loss associated with anti-retinal and anti-optic nerve antibodies while on maintenance nivolumab immunotherapy. Case 3 developed decreased visual acuity due to acute exudative polymorphous vitelliform maculopathy within 2 weeks of initiating ipilimumab/nivolumab immunotherapy. All patients had concurrent extraocular immune-related adverse events in addition to the presence of anti-retinal antibodies on serological testing. 14 published cases of AIR associated with immunotherapy for cutaneous or non-ocular mucosal melanoma were identified and reviewed. CONCLUSIONS: Immune checkpoint inhibition can trigger the development of AIR with varied clinical manifestations in patients with advanced cutaneous melanoma. This study highlights the need for close monitoring in cutaneous melanoma patients receiving immunotherapy who develop new visual symptoms with or without funduscopic changes, as well as the potential role for screening of patients prior to initiating immunotherapy.

Ethnicity-based name partitioning for author name disambiguation using supervised machine learning.

In several author name disambiguation studies, some ethnic name groups such as East Asian names are reported to be more difficult to disambiguate than others. This implies that disambiguation approaches might be improved if ethnic name groups are distinguished before disambiguation. We explore the potential of ethnic name partitioning by comparing performance of four machine learning algorithms trained and tested on the entire data or specifically on individual name groups. Results show that ethnicity-based name partitioning can substantially improve disambiguation performance because the individual models are better suited for their respective name group. The improvements occur across all ethnic name groups with different magnitudes. Performance gains in predicting matched name pairs outweigh losses in predicting nonmatched pairs. Feature (e.g., coauthor name) similarities of name pairs vary across ethnic name groups. Such differences may enable the development of ethnicity-specific feature weights to improve prediction for specific ethic name categories. These findings are observed for three labeled data with a natural distribution of problem sizes as well as one in which all ethnic name groups are controlled for the same sizes of ambiguous names. This study is expected to motive scholars to group author names based on ethnicity prior to disambiguation.

Dabrafenib and Trametinib for BRAF-Mutated Conjunctival Melanoma.

BACKGROUND: Conjunctival melanoma is a potentially lethal malignancy of the ocular surface. There have been no therapeutic advancements made in the past several decades despite increasing prevalence of the disease. METHODS: The authors report the case of a 52-year-old Caucasian male with unresectable, recurrent conjunctival melanoma with V600 BRAF mutation who was treated with systemic BRAF/MEK inhibition. RESULTS: There was complete regression of local disease within the first 9 months. The patient remains without local recurrence or systemic metastasis at 1 year. CONCLUSION: This is the first reported case of conjunctival melanoma with complete response to BRAF/MEK inhibition. As long as targeted therapy remains an option, patients with conjunctival melanoma should undergo mutational profiling of their tumor.

Chondrosarcoma Metastasis to the Choroid.

We report a rare case of chondrosarcoma metastatic to the choroid. A 64-year-old male with a history of chondrosarcoma metastatic to the lungs and to the spine presented with blurred vision. A choroidal tumor was found. Fine-needle biopsy confirmed the histologic identity of the tumor as chondrosarcoma. Metastatic spread of chondrosarcoma to the eye is extremely rare. When present, lesions may grow rapidly, and systemic prognosis is poor. Co-management with medical oncology is of utmost importance. This is the third case of chondrosarcoma metastatic to the choroid in the literature and the first with bilateral involvement.

A Focused Review of Smartphone Diet-Tracking Apps: Usability, Functionality, Coherence With Behavior Change Theory, and Comparative Validity of Nutrient Intake and Energy Estimates.

BACKGROUND: Smartphone diet-tracking apps may help individuals lose weight, manage chronic conditions, and understand dietary patterns; however, the usabilities and functionalities of these apps have not been well studied. OBJECTIVE: The aim of this study was to review the usability of current iPhone operating system (iOS) and Android diet-tracking apps, the degree to which app features align with behavior change constructs, and to assess variations between apps in nutrient coding. METHODS: The top 7 diet-tracking apps were identified from the iOS iTunes and Android Play online stores, downloaded and used over a 2-week period. Each app was independently scored by researchers using the System Usability Scale (SUS), and features were compared with the domains in an integrated behavior change theory framework: the Theoretical Domains Framework. An estimated 3-day food diary was completed using each app, and food items were entered into the United States Department of Agriculture (USDA) Food Composition Databases to evaluate their differences in nutrient data against the USDA reference. RESULTS: Of the apps that were reviewed, LifeSum had the highest average SUS score of 89.2, whereas MyDietCoach had the lowest SUS score of 46.7. Some variations in features were noted between Android and iOS versions of the same apps, mainly for MyDietCoach, which affected the SUS score. App features varied considerably, yet all of the apps had features consistent with Beliefs about Capabilities and thus have the potential to promote self-efficacy by helping individuals track their diet and progress toward goals. None of the apps allowed for tracking of emotional factors that may be associated with diet patterns. The presence of behavior change domain features tended to be weakly correlated with greater usability, with R2 ranging from 0 to .396. The exception to this was features related to the Reinforcement domain, which were correlated with less usability. Comparing the apps with the USDA reference for a 3-day diet, the average differences were 1.4% for calories, 1.0% for carbohydrates, 10.4% for protein, and -6.5% for fat. CONCLUSIONS: Almost all reviewed diet-tracking apps scored well with respect to usability, used a variety of behavior change constructs, and accurately coded calories and carbohydrates, allowing them to play a potential role in dietary intervention studies.

Novel Eye Plaque Designs for Brachytherapy of Iris and Ciliary Body Melanoma and the First Clinical Application.

BACKGROUND: While traditional eye plaque brachytherapy can be used for the treatment of iris melanoma, it faces challenges of poor patient tolerability due to cornea-plaque touch caused by radius of curvature mismatch and potential dosimetric inaccuracy from incomplete coverage. We present novel plaque designs and the first clinical application of the plaques for iris melanoma. METHODS: Two dome-shaped plaques (EP2132 and EP1930) were designed to vault above the cornea to treat tumors of the iris and ciliary body. Image-based treatment planning of the first 2 clinical cases using the EP2132 plaque covered the tumor base plus a 2 mm margin and the involved ciliary body with at least 75 Gy to the tumor apex. RESULTS: The tumors decreased in size following treatment. The patients tolerated the treatment well. There was no adverse event associated with the traditional iris plaques, such as decreased vision, pain, corneal edema, glaucoma, or cataract. CONCLUSION: The novel dome-shaped plaques for the treatment of iris melanoma provide effective dose distribution, improved surgical maneuverability, and increased tolerability for the patient. This plaque model can be used to treat iris melanoma of various sizes, configurations, and locations, including the ciliary body. The need for a customized plaque platform for each patient is minimized.

Spinal Ewing Sarcoma Metastatic to Bilateral Orbits.

A 39-year-old woman with primary spinal Ewing sarcoma and known lung metastases presented with painless bilateral decreased visual acuity over a 1-month period. Examination revealed bilateral disc edema. MRI of the brain/orbits showed metastatic lesions to the dura and bilateral orbits. Venous sinus thrombosis extending to the jugular vein was also noted. To the authors' knowledge, this is the first case of bilateral orbital metastasis and papilledema secondary to Ewing sarcoma and related hypercoagulability.

Orbital cellulitis with choroidal detachment following strabismus surgery in an adult.

We report the case of a 76-year-old man who presented with left eye pain, eyelid swelling, and restricted extraocular motility 3 days after strabismus surgery. Computed tomography of the orbits did not show drainable abscesses, but magnetic resonance imaging on postoperative day 9 showed choroidal detachment with retrobulbar abscess and dural enhancement despite broad-spectrum intravenous antibiotic use. Following incision and drainage of the abscess, which grew methicillin-resistant Staphylococcus aureus, the patient showed improvement in visual acuity, pain, and extraocular motility. This is the first report of choroidal detachment occurring in the setting of orbital cellulitis after strabismus surgery.

Incorporating patient-specific CT-based ophthalmic anatomy in modeling iodine-125 eye plaque brachytherapy dose distributions.

PURPOSE: To quantify the dosimetric impact of incorporating patient-specific CT-based models rather than the conventional stylized-standard model for eye plaque brachytherapy planning. METHODS AND MATERIALS: Plaque Simulator was used to plan 16 patients using both CT-based patient-specific eye model and stylized-standard (SS) eye models. Plaque position was initially based on the SS model and compared against their patient-specific model without changing the plaque loading pattern and seed strength. Dosimetric parameters were compared for tumor and healthy ocular structures. RESULTS: Patient-specific ocular parameters ranged from 0.40 to 1.38 of SS model values. If plaques were placed based on SS model eyelet positions, target volume receiving prescription dose (V100%) is overpredicted by 5.9% on average (max: 27%), and D95% is overpredicted by 17.2 Gy on average (max: 58.1 Gy). If the plaques were recentered, 13 of 16 patients had changes in V100% of less than 2%, whereas half of the patients still had optic disc dose difference greater than 5 Gy (max: 36.2 Gy). The largest differences were observed with a target-to-optic disk distance less than 6 mm. No substantial dose differences were observed for the tumor apex, fovea, lens, and opposing retina. CONCLUSIONS: Patient-specific modeling is recommended for clinical planning, especially with target-to-optic disk distances less than 6 mm, due to significant differences compared with SS model.

Blepharoplasty markers: comparison of ink drying time and ink spread.

Marking of the eyelid is a crucial presurgical step in blepharoplasty. A number of markers are available for this purpose with variable ink characteristics. In this study, we measured the ink drying time and spread width of 13 markers used for preoperative marking for blepharoplasty. Based on the results, we propose markers that may be best suited for use in this procedure.

Acute Presentation of Mesectodermal Leiomyoma of the Ciliary Body.

PURPOSE: We report a case of acutely presenting mesectodermal leiomyoma of the ciliary body in a 29-year-old female who reported waking up with swollen eyelids of the right eye and light-perception vision. The affected eye had elevated intraocular pressure, a flat anterior chamber, and a pale, round mass arising from the nasal ciliary body, invading the angle and protruding into the visual axis posterior to the lens. Within days, the visual acuity decreased to no light perception. The eye was enucleated. METHODS: The enucleated eye harbored a tumor arising from the ciliary body, measuring 18 mm in the greatest dimension. Spindled cells with fibrillary cytoplasmic processes suggested a neural origin though negative for S-100, Melan-A, and HMB-45. The cells stained strongly positive for smooth muscle actin and vimentin, leading to the diagnosis of mesectodermal leiomyoma of the ciliary body. RESULTS: We review the literature to expand upon the clinical findings, diagnostic methods, and histopathologic and immunohistochemistry characteristics of mesectodermal leiomyoma. CONCLUSION: Leiomyoma must be in the differential diagnosis for ciliary body mass, especially in women of reproductive age. Diagnosis relies on histopathology and immunohistochemistry. The mechanism of acute symptom onset may be multifactorial. This case emphasizes the possibility of acute presentation of a rare, benign intraocular tumor.

Effect of endoscopic cyclophotocoagulation on refractive outcomes when combined with cataract surgery.

OBJECTIVE: To determine the difference between predicted and actual refractive outcomes after combined endoscopic cyclophotocoagulation and cataract surgery (phaco-ECP) in patients with open-angle glaucoma (OAG). DESIGN: Retrospective case-control study of patients with OAG who underwent phaco-ECP compared with cataract surgery alone. PARTICIPANTS: Eighty-three patients with OAG, aged 55 to 91 years, who underwent a combined phaco-ECP procedure and 58 biometry- and age-matched control patients with OAG who underwent cataract surgery alone. METHODS: Patient records were retrospectively reviewed at the Department of Ophthalmology and Visual Sciences, Washington University School of Medicine in St. Louis (St. Louis, Mo.). The primary outcome was the difference in predicted and actual refractive outcomes in patients undergoing either phaco-ECP or standard cataract surgery. RESULTS: Compared with phaco alone, the difference in predicted versus actual postoperative results was more myopic in the phaco-ECP group (0.029, -0.110, and -0.095 vs -0.169, -0.325, and -0.312 [p < 0.05] for Sanders, Retzlaff, Kraff/Theoretical, Hoffer Q, and Holladay, respectively). Moreover, the F test for variability showed significantly more variability in refractive outcomes in the phaco-ECP group compared with standard cataract surgery. CONCLUSIONS: Patients undergoing phaco-ECP may have postoperative refractive errors that may vary from that predicted preoperatively more so than in cataract surgery alone. Surgeons may consider analyzing their results to determine whether any adjustment should be made to lens power selection when performing phaco-ECP.

The involvement of the orbitofrontal cortex in learning under changing task contingencies.

Previous investigations examining the rat prefrontal cortex subregions in attentional-set shifting have commonly employed two-choice discriminations. To better understand how varying levels of difficulty influence the contribution of the prefrontal cortex to learning, the present studies examined the effects of orbitofrontal cortex inactivation in a two- or four-choice odor reversal learning test. Long-Evans rats were trained to dig in cups that contained distinct odors. In the two-choice odor discrimination, one odor cup was always associated with a cereal reinforcement in acquisition while the opposite odor cup was associated with a cereal reinforcement in reversal learning. In the four-choice odor discrimination, an additional two cups containing distinct odors were used that were never associated with reinforcement in acquisition or reversal learning. Bilateral infusions of the GABA-A agonist, muscimol (0.5 microg) into the orbitofrontal cortex did not impair acquisition of either the two- or four-choice discrimination task. However, muscimol infusions into the orbitofrontal cortex impaired two- and four-choice reversal learning. In the two-choice odor reversal, muscimol treatment selectively increased perseverative errors. In the four-choice odor reversal, muscimol treatment increased perseverative, regressive, as well as irrelevant errors. These findings suggest that the orbital prefrontal cortex not only enables task switching by supporting the initial inhibition of a previously relevant choice pattern, but under increasing task demands also enables the reliable execution of a new choice pattern and reduction of interference to irrelevant stimuli.

The contribution of the rat prelimbic-infralimbic areas to different forms of task switching.

The experiments examined the effects of prelimbic-infralimbic inactivation in rats on the acquisition and reversal learning of different discrimination tasks: 2- or 4-choice odor discrimination in Experiments 1 and 2, the shift from 2-choice odor discrimination to 2-choice place discrimination in Experiment 3, and the shift from 2-choice place to 2-choice odor discrimination in Experiment 4. Infusions of 2% bupivacaine did not impair performance in the odor discrimination tests. Prelimbic-infralimbic inactivation did not impair acquisition but did impair the shift from an odor to a place discrimination and vice versa. Analysis of the errors revealed that the deficit was due to perseveration of the previously learned strategy. The selective deficits observed in the odor-place tests suggest that the prelimbic-infralimbic areas enable behavioral flexibility when conditions demand inhibiting the use of one type of attribute information and learning a new type of attribute information.