Automatic Hip Fracture Identification and Functional Subclassification with Deep Learning
Authors:
Justin D Krogue,
Kaiyang V Cheng,
Kevin M Hwang,
Paul Toogood,
Eric G Meinberg,
Erik J Geiger,
Musa Zaid,
Kevin C McGill,
Rina Patel,
Jae Ho Sohn,
Alexandra Wright,
Bryan F Darger,
Kevin A Padrez,
Eugene Ozhinsky,
Sharmila Majumdar,
Valentina Pedoia
Abstract:
Purpose: Hip fractures are a common cause of morbidity and mortality. Automatic identification and classification of hip fractures using deep learning may improve outcomes by reducing diagnostic errors and decreasing time to operation. Methods: Hip and pelvic radiographs from 1118 studies were reviewed and 3034 hips were labeled via bounding boxes and classified as normal, displaced femoral neck f…
▽ More
Purpose: Hip fractures are a common cause of morbidity and mortality. Automatic identification and classification of hip fractures using deep learning may improve outcomes by reducing diagnostic errors and decreasing time to operation. Methods: Hip and pelvic radiographs from 1118 studies were reviewed and 3034 hips were labeled via bounding boxes and classified as normal, displaced femoral neck fracture, nondisplaced femoral neck fracture, intertrochanteric fracture, previous ORIF, or previous arthroplasty. A deep learning-based object detection model was trained to automate the placement of the bounding boxes. A Densely Connected Convolutional Neural Network (DenseNet) was trained on a subset of the bounding box images, and its performance evaluated on a held out test set and by comparison on a 100-image subset to two groups of human observers: fellowship-trained radiologists and orthopaedists, and senior residents in emergency medicine, radiology, and orthopaedics. Results: The binary accuracy for fracture of our model was 93.8% (95% CI, 91.3-95.8%), with sensitivity of 92.7% (95% CI, 88.7-95.6%), and specificity 95.0% (95% CI, 91.5-97.3%). Multiclass classification accuracy was 90.4% (95% CI, 87.4-92.9%). When compared to human observers, our model achieved at least expert-level classification under all conditions. Additionally, when the model was used as an aid, human performance improved, with aided resident performance approximating unaided fellowship-trained expert performance. Conclusions: Our deep learning model identified and classified hip fractures with at least expert-level accuracy, and when used as an aid improved human performance, with aided resident performance approximating that of unaided fellowship-trained attendings.
△ Less
Submitted 10 September, 2019;
originally announced September 2019.
Efficient Hybrid White Organic Light-Emitting Diodes for Application of Triplet Harvesting with Simple Structure
Authors:
Kyo Min Hwang,
Song Eun Lee,
Sungkyu Lee,
Han Kyu Yoo,
Hyun Jung Baek,
Jwajin Kim,
Seung Soo Yoon,
Young Kwan Kim
Abstract:
In this study, we fabricated hybrid white organic light-emitting diodes (WOLEDs) based on triplet harvesting with simple structure. All the hole transporting material and host in emitting layer (EML) of devices were utilized with same material by using N,N'-di-1-naphthalenyl-N,N'-diphenyl-[1,1':4',1":4",1"'-quaterphenyl]-4,4"'-diamine (4P-NPD) which were known to be blue fluorescent material. Simp…
▽ More
In this study, we fabricated hybrid white organic light-emitting diodes (WOLEDs) based on triplet harvesting with simple structure. All the hole transporting material and host in emitting layer (EML) of devices were utilized with same material by using N,N'-di-1-naphthalenyl-N,N'-diphenyl-[1,1':4',1":4",1"'-quaterphenyl]-4,4"'-diamine (4P-NPD) which were known to be blue fluorescent material. Simple hybrid WOLEDs were fabricated three color with blue fluorescent and green, red phosphorescent materials. We was investigated the effect of triplet harvesting (TH) by exciton generation zone on simple hybrid WOLEDs. Characteristic of simple hybrid WOLEDs were dominant hole mobility, therefore exciton generation zone was expected in EML. Additionally, we was optimization thickness of hole transporting layer and electron transporting layer was fabricated a simple hybrid WOLEDs. Simple hybrid WOLED exhibits maximum luminous efficiency of 29.3 cd/A and maximum external quantum efficiency of 11.2%. Commission Internationale de l'Eclairage (International Commission on Illumination) coordinates of (0.45, 0.43) at about 10,000 cd/m^2.
△ Less
Submitted 23 May, 2016;
originally announced May 2016.
