992 research outputs found
Film Review: Yoo Hoo, Mrs. Goldberg
No full textThe author presents a review of the documentary Yoo Hoo, Mrs. Goldberg
Data for "Prediction of Phakic Intraocular Lens Vault Using Machine Learning of Anterior Segment Optical Coherence Tomography Metrics"
No full textPrediction of Phakic Intraocular Lens Vault Using Machine Learning of Anterior Segment Optical Coherence Tomography Metrics.
Authors: Kazutaka Kamiya, MD, PhD1, Ik Hee Ryu, MD, MS2, Tae Keun Yoo, MD2, Jung Sub Kim MD2, In Sik Lee, MD, PhD2, Jin Kook Kim MD2, Wakako Ando CO3, Nobuyuki Shoji, MD, PhD3, Tomofusa, Yamauchi, MD, PhD4, Hitoshi Tabuchi, MD, PhD4.
Author Affiliation: 1Visual Physiology, School of Allied Health Sciences, Kitasato University, Kanagawa, Japan, 2B&VIIT Eye Center, Seoul, Korea, 3Department of Ophthalmology, School of Medicine, Kitasato University, Kanagawa, Japan, 4Department of Ophthalmology, Tsukazaki Hospital, Hyogo, Japan.
We hypothesize that machine learning of preoperative biometric data obtained by the As-OCT may be clinically beneficial for predicting the actual ICL vault. Therefore, we built the machine learning model using Random Forest to predict ICL vault after surgery.
This multicenter study comprised one thousand seven hundred forty-five eyes of 1745 consecutive patients (656 men and 1089 women), who underwent EVO ICL implantation (V4c and V5 Visian ICL with KS-AquaPORT) for the correction of moderate to high myopia and myopic astigmatism, and who completed at least a 1-month follow-up, at Kitasato University Hospital (Kanagawa, Japan), or at B&VIIT Eye Center (Seoul, Korea).
This data file (RFR_model(feature=12).mat) is the final trained random forest model for MATLAB 2020a.
Python version:
***************************************************************
from sklearn.model_selection import train_test_split
import pandas as pd
import numpy as np
from sklearn.ensemble import RandomForestClassifier
from sklearn.ensemble import RandomForestRegressor
# connect data in your google drive
from google.colab import auth
auth.authenticate_user()
from google.colab import drive
drive.mount('/content/gdrive')
# Change the path for the custom data
# In this case, we used ICL vault prediction using preop measurement
dataset = pd.read_csv('gdrive/My Drive/ICL/data_icl.csv')
dataset.head()
#optimal features (sorted by importance) :
# 1. ICL size 2. ICL power 3. LV 4. CLR 5. ACD 6. ATA
# 7. MSE 8.Age 9. Pupil size 10. WTW 11. CCT 12. ACW
y = dataset['Vault_1M']
X = dataset.drop(['Vault_1M'], axis = 1)
# Split the dataset to train and test data
# For a simple validation test, we split data to 8:2
train_X, test_X, train_y, test_y = train_test_split(X, y, test_size=0.2, random_state=0)
# Optimal parameter search could be performed in this section
parameters = {'bootstrap': True,
'min_samples_leaf': 3,
'n_estimators': 500,
'criterion': 'mae'
'min_samples_split': 10,
'max_features': 'sqrt',
'max_depth': 6,
'max_leaf_nodes': None}
RF_model = RandomForestRegressor(**parameters)
RF_model.fit(train_X, train_y)
RF_predictions = RF_model.predict(test_X)
importance = RF_model.feature_importances
Young Investigator: Michelle J Yoo
No full text Supervisor’s supporting comments I have always been impressed with Michelle’s ability to conduct research in an independent and yet highly effective manner. Part of her research in my group has examined the use of affinity columns to examine drug–protein binding with serum proteins, such as human serum albumin. This work is extremely important to the fields of pharmaceutical chemistry and clinical chemistry in providing the data needed for the development of new drugs or in the optimization of treatments for patients with new, or existing, drugs. Another topic that Michelle has examined in her research is the use of new supports based on monolithic materials and ultrafast-extraction methods for affinity-based separations of biological samples and high-throughput screening of drug–protein binding. She was the lead author on a review written on this topic and also has several research publications related to this area of work. During her graduate studies, Michelle has emerged as a real leader in my group. She has excellent people and communication skills and is highly motivated in her pursuit of an advanced degree in analytical chemistry and bioanalysis. I have extremely high expectations for her in the future as she continues her career. Nominated by: David S Hage, University of Nebraska, Department of Chemistry, Hamilton Hall 704, Lincoln, NE 68588, USA </jats:p
Urban climate characterization and heat risk assessment based on machine learning and remote sensing data
No full textDepartment of Urban and Environmental Engineering (Environmental Science and Engineering)clos
All-Sky 1 km MODIS Land Surface Temperature Reconstruction Considering Cloud Effects Based on Machine Learning
No full textOpen AccessArticle
All-Sky 1 km MODIS Land Surface Temperature Reconstruction Considering Cloud Effects Based on Machine Learning
by Dongjin ChoORCID,Dukwon Bae,Cheolhee YooORCID,Jungho Im *ORCID,Yeonsu LeeORCID andSiwoo LeeORCID
Department of Urban and Environmental Engineering, Ulsan National Institute of Science and Technology (UNIST), Ulsan 44919, Korea
*
Author to whom correspondence should be addressed.
Academic Editor: Anand Inamdar
Remote Sens. 2022, 14(8), 1815; https://doi.org/10.3390/rs14081815
Received: 9 February 2022 / Revised: 5 April 2022 / Accepted: 7 April 2022 / Published: 9 April 2022
(This article belongs to the Special Issue Land Surface Temperature Estimation Using Remote Sensing)
Download PDF Browse Figures Citation Export
Abstract
A high spatio-temporal resolution land surface temperature (LST) is necessary for various research fields because LST plays a crucial role in the energy exchange between the atmosphere and the ground surface. The moderate-resolution imaging spectroradiometer (MODIS) LST has been widely used, but it is not available under cloudy conditions. This study proposed a novel approach for reconstructing all-sky 1 km MODIS LST in South Korea during the summer seasons using various data sources, considering the cloud effects on LST. In South Korea, a Local Data Assimilation and Prediction System (LDAPS) with a relatively high spatial resolution of 1.5 km has been operated since 2013. The LDAPS model???s analysis data, binary MODIS cloud cover, and auxiliary data were used as input variables, while MODIS LST and cloudy-sky in situ LST were used together as target variables based on the light gradient boosting machine (LightGBM) approach. As a result of spatial five-fold cross-validation using MODIS LST, the proposed model had a coefficient of determination (R2) of 0.89???0.91 with a root mean square error (RMSE) of 1.11???1.39 ??C during the daytime, and an R2 of 0.96???0.97 with an RMSE of 0.59???0.60 ??C at nighttime. In addition, the reconstructed LST under the cloud was evaluated using leave-one-station-out cross-validation (LOSOCV) using 22 weather stations. From the LOSOCV results under cloudy conditions, the proposed LightGBM model had an R2 of 0.55???0.63 with an RMSE of 2.41???3.00 ??C during the daytime, and an R2 of 0.70???0.74 with an RMSE of 1.31???1.36 ??C at nighttime. These results indicated that the reconstructed LST has higher accuracy than the LDAPS model. This study also demonstrated that cloud cover information improved the cloudy-sky LST estimation accuracy by adequately reflecting the heterogeneity of the relationship between LST and input variables under clear and cloudy skies. The reconstructed all-sky LST can be used in a variety of research applications including weather monitoring and forecasting
Accessing and using multilanguage information by users searching in differenct information retrieval systems
No full textThere is an underlying assumption in the exchange of scholarly information that knowledge will be transferred across country borders, cultures, and languages. It is this sharing of scholarly information is considered an essential pre-requisite necessary for the advancement of knowledge. Nonetheless, in the current English dominant environment of information retrieval (IR) systems, there are numerous obstacles confronting users who seek to access and use non-English information.
The purposes of this study are: to explore the information behaviors of those seeking non-English information; to identify difficulties of individuals' experiences when accessing and using non-English information in current IR systems; to develop an explanatory model determining how person characteristics, experiential knowledge, and situation factors influence search behaviors and evaluations of bibliographic information.
Two separate studies are conducted to explore the above issues: an online questionnaire of users of multilanguage information retrieval systems; and an experiment with individuals accessing information on different topics using different languages and systems. The participants in these studies include academic researchers and library personnel and are individuals who regularly interact with Chinese, Japanese, Korean and English records via IR systems.
The survey and experiment participants note the lack of non-English access via indexing terms, the lack of non-English records in major online databases which index journals, the lack of English translation of abstracts, and the lack of coherent and understandable access to non-Roman language materials. The users of non-English information expect to have a system with cross language information retrieval functions providing clear access to full text non-English information. Importantly, having understandable bibliographic records are essential when individuals make decisions on their expected use of non-English documents.
The experiment data analyses reveal there are different IR system search behaviors by subjects' with different language backgrounds, professions, language knowledge, topic knowledge and its target language, especially comparing English with non-English searches. An explanatory model for non-English searching model was built based on various statistical analyses of experiment data. The model depicts the importance of statistically significant relationships among person characteristics and experiential knowledge which explain search behaviors and intention to use retrieved information when individuals seek non-English/non-Roman alphabet information.Ph.D.Includes bibliographical references (p. 226-238)by Yoo Jin H
Kochosa sharae Framenau & Castanheira & Yoo 2023, sp. nov.
No full textKochosa sharae sp. nov. (Figs 20, 22A–E) Holotype. Male, Flinders Chase National Park, 4 km W Rocky River Headquarters, Kangaroo Island (35º 57'00''S 136º42'30''E, South Australia, AUSTRALIA), E. G. Matthews, J. A. Forrest, 1–7 November 1990 (SAM NN13502). Etymology. The specific epithet is a matronym honouring a good friend of the senior author, Shar Ramamurthy, currently Senior Manager, Environmental Water at the Department of Environment, Land, Water and Planning (Victoria), for her support during the mutual times at Melbourne University. Other material examined. Australia: South Australia: 2 males, Snug Cove, 9 km NNE, Kangaroo Island, 35º47'19''S 136º48'59''E (NN13503–4). Diagnosis. Male pedipalp morphology of K. sharae sp. nov. is similar to that of K. fleurae sp. nov., both sharing a digitiform basoembolic apophysis; however, it is comparatively smaller and more curved in K. sharae sp. nov. (Fig. 22E). Description. Male (based on holotype, SAM NN13502). Cephalothorax. Dorsally dark brown; broad median light band narrowing posteriorly; lateral light bands distinct; white setae throughout (Fig. 22A). Sternum dark brown with few white setae (Fig. 22B). Abdomen. Dorsally dark olive brown; cardiac mark continuous, narrowest posteriorly and there bordered by black spots (Fig. 22A). Venter light olive-brown, anteriorly somewhat darker (Fig. 22B). Pedipalps (Figs 22C–E). Patella of distinct light brown colouration with white setae; tegular apophysis round lobe, almost translucent; basoembolic apophysis digitiform and curved; embolus straight medially, curved basally and then apically. (Fig. 22E) Legs. Brown with darker annulations; spination of leg I: femur. 2 dorsum, 1 apicodorsal, 1 apicoprolateral, 1 apicoretrolateral (very small); tibia. 3 ventral pairs, 2 prolateral, 2 retrolateral; metatarsus. 3 ventral pairs, 1 apicoventral, 2 prolateral, 1 apicoprolateral, 2 retrolateral, 1 apicoretrolateral. Measurements: TL 4.84, CL 2.67, CW 1.99. Eyes: AME 0.10, ALE 0.10, PME 0.24, PLE 0.23. Row of eyes: AE 0.64, PME 0.76, PLE 0.96. Sternum (length/width) 1.21/0.99. Labium (length/width) 0.36/0.38. AL 2.17, AW 1.49. Legs: Length of segments: Pedipalp 1.00+0.91+-+0.81=2.72, I 1.80+2.01+1.44+0.96=6.21; II 1.68+2.04+1.4 6+0.96=6.14, III 1.68+1.84+1.56+0.91=5.99; IV 2.06+2.32+2.30+1.24=7.92. Variation. Size (range, mean ± s.d.): TL 4.60–4.84, 4.75 ± 0.13; CL 2.65–2.72, 2.68 ± 0.04; CW 1.85–1.99, 1.91 ± 0.07, n = 3. One of the males was much darker than the one illustrated here and with less distinct median carapace band and abdominal cardiac mark. Female. Unknown. Life history and habitat preferences. There is no information on the habitat of this species; the three males of K. sharae sp. nov. were found between October to November suggesting this to be a spring-mature species. Distribution. Kochosa sharae sp. nov. is only known from western Kangaroo Island, South Australia (Fig. 20).Published as part of Framenau, Volker W., Castanheira, Pedro De S. & Yoo, Jung-Sun, 2023, The artoriine wolf spiders of Australia: the new genus Kochosa and a key to genera (Araneae: Lycosidae), pp. 301-357 in Zootaxa 5239 (3) on page 338, DOI: 10.11646/zootaxa.5239.3.1, http://zenodo.org/record/763479
Estimation of maximum and minimum air temperatures in urban areas using MODIS satellite data
No full textKochosa tongiorgii Framenau & Castanheira & Yoo 2023, sp. nov.
No full textKochosa tongiorgii sp. nov. (Figs 27, 30A–E) Holotype. 1 male, Davies Creek National Park (17º00'S 145º34'E, Queensland, AUSTRLIA), 26 November 1992 – 15 April 1993, R. J. & S. Raven, P. & E. Lawless (QM S83707). Etymology. The specific epithet is a patronym for the late Paolo Tongiorgi (1936–2018) for his contribution to wolf spider taxonomy. The senior author has fond memories of Paolo’s support during his work on Arctosa cinerea Fabricius, 1777 (e.g., Framenau 1995). Other material examined. 4 males, 1 juvenile, same data as holotype (QM S19781). Diagnosis. Males of K. tongiorgii sp. nov. differ from all other species within the genus by the unique colouration of the carapace; dense white setae cover it almost entirely except for the flanks of the cephalic area (Fig. 30A). The long and tapering embolus is opposed by a sclerotised, digitiform process originating at the retrolateral edge of the apical division, unique within the genus (Fig. 30E). The female of K. tongiorgii sp. nov. is unknown. Description Male (based on holotype, QM S83707). Cephalothorax. Dorsally dark brown, covered almost entirely (except cephalic flanks) by white, short and stout setae (Fig. 30A). Sternum brown (Fig. 30B). Abdomen. Dorsally dark brown with continuous cardiac mark, dense cover of white shot and stout setae obscure colouration as on carapace (Fig. 30A). Venter olive-grey, centrally somewhat lighter (Fig. 30B). Pedipalps (Fig. 30C–E). Embolic base exposed; tegular apophysis broadly subtriangular; embolus narrow and long; apical part of embolus supported by trough-shaped structure; sclerotised process opposing embolus. Legs. Brown, femora darkest; spination of leg I: femur: 3 dorsal; tibia: 3 ventral pairs and 1 apicoventral; metatarsus: 4 ventral pairs, 1 prolateral, 1 apicoprolateral. Measurements. TL 3.46, CL 1.93, CW 1.25. Eyes: AME 0.07, ALE 0.06, PME 0.17, PLE 0.17. Row of eyes: AE 0.47, PME 0.63, PLE 0.70. Sternum (length/width) 0.81/0.72. Labium (length/width) 0.24/0.28. AL 1.62, AW 0.94. Legs: Length of segments: Pedipalp 0.71+0.67+-+0.64=2.02, I 1.21+1.60+1.03+0.74=4.58; II 1.21+1.49+1.0 3+0.71=4.44, III 1.10+1.28+1.14+0.64=4.16; IV 1.67+1.92+1.82+0.89=6.30. Variation. Size (range, mean ± s.d.): TL 3.46–3.71, 3.57 ± 0.15; CL 1.93–2.20, 2.06 ± 0.10; CW 1.25–1.40, 1.30 ± 0.06, n = 5. There was not major difference in colour pattern in any of the males, although the abdomen was somewhat lighter in one specimen. Female. Unknown. Life history and habitat preferences. Unknown. Distribution. Kochosa tongiorgii sp. nov. is only know from the type locality, Davies Creek National Park, north-eastern Queensland (Fig. 27).Published as part of Framenau, Volker W., Castanheira, Pedro De S. & Yoo, Jung-Sun, 2023, The artoriine wolf spiders of Australia: the new genus Kochosa and a key to genera (Araneae: Lycosidae), pp. 301-357 in Zootaxa 5239 (3) on page 350, DOI: 10.11646/zootaxa.5239.3.1, http://zenodo.org/record/763479
Kochosa fleurae Framenau & Castanheira & Yoo 2023, sp. nov.
No full textKochosa fleurae sp. nov. (Figs 13, 14A–E) Holotype. Male, Skylark Road, Whipstick (36º37'30''S 144º16'00″E, Victoria, AUSTRALIA), J. Shield, 10–17 December 1998, Whipstick Forest Survey MS # 138 (MV K-8672). Etymology. The specific epithet is a matronym for a very good friend of the senior author, Fleur de Crespigny (currently Head, Dementia Unit at the Australian Institute of Health and Welfare, Canberra). Other material examined. 2 males, Skylark Road, Whipstick, 36º37'30''S 144º16'00″E (MV K-7728). Diagnosis. Males of K. fleurae sp. nov. are separated from those of all other species of the genus by their unique long and sharp basoembolic apophyses and flat and broad embolus (Fig. 14E). They are most similar to K. asterix sp. nov. but differ from those due to the ventrally folded basoembolic apophysis (Fig. 14E). The female of K. fleurae sp. nov. is unknown. Description. Male (based on holotype, MV K-8672). Cephalothorax. Dorsally dark brown; median light band slightly narrowing posteriorly, covered with white setae; lateral light bands broad and covered with white setae (Fig. 14A). Sternum dark brown (Fig. 14B). Abdomen. Dorsally light olive grey; cardiac mark continuous bordered by indistinct dark brown discolourations in posterior half (Fig. 14A). Venter brown (Fig. 14B). Pedipalps (Fig. 14C–E). Embolus broad and flat and tapering to apex; basoembolic apophysis distinct, long and pointing retrolaterally; tegular apophysis forms a small round lobe situated apically on tegulum. Legs. Light brown with irregular darker discolourations; spination of leg I: femur: 4 dorsal, 2 retrolateral (very small); tibia: 4 ventral pairs (apical pair small), 2 prolateral, 1 retrolateral; metatarsus: 4 ventral pairs (apical pair small and closer), 2 prolateral; 1 apicoprolateral, 1 retrolateral. Measurements. TL 4.39, CL 2.49, CW 1.64. Eyes: AME 0.10, ALE 0.10, PME 0.31, PLE 0.30. Row of eyes: AE 0.48, PME 0.65, PLE 0.79. Sternum (length/width) 0.99/0.92. Labium (length/width) 0.28/0.34. AL 1.99, AW 1.35. Legs: Length of segments: Pedipalp 0.84+0.28+0.36+0.88 = 2.36, I 1.42+1.83+1.17+0.85=5.27; II 1.42+1.78 +1.17+0.89=5.26, III 1.35+1.49+1.24+0.74=4.82; IV 1.85+2.24+2.14+1.07=7.30. Variation. Size (range, mean ± s.d.): TL 4.39 – 5.82, 4.93 ± 0.75; CL 2.49 – 3.18, 2.78 ± 0.35; CW 1.64–1.91, 1.75 ± 0.14, n = 3. Both other males are of similar colouration as the holotype. Female. Unknown. Life history and habitat preferences. The three males of K. fleurae sp. nov. were found in Whipstick Westringia (Westringia crassifolia) shrub and forest between October and December. Westringia crassifolia is listed as Endangered in Victoria and Australia and this conservation rating may also be applied to K. fleurae sp. nov. if the species is a habitat specialist relying on this species. Distribution. Only known from and around the type locality, Whipstick (Victoria) (Fig. 13).Published as part of Framenau, Volker W., Castanheira, Pedro De S. & Yoo, Jung-Sun, 2023, The artoriine wolf spiders of Australia: the new genus Kochosa and a key to genera (Araneae: Lycosidae), pp. 301-357 in Zootaxa 5239 (3) on pages 325-326, DOI: 10.11646/zootaxa.5239.3.1, http://zenodo.org/record/763479
- …
