Bone X-Rays Classification and Abnormality Detection

This paper introduces a novel automatic reliable hybrid two-stage method for bone x-rays abnormality detection. For this purpose, 10 different pre-trained convolutional neural networks architectures with different number of layers are examined. The introduced method considers the seven extremity upper bones, namely shoulder, humerus, forearm, elbow, wrist, hand, and finger. The enhanced images are fed into the first stage to classify the bone type into one of the seven bones. Thereafter, the abnormality is detected in the second stage using a specific classifier according to the bone type. Thus, the classification step consists of eight different classifiers: one for the bone classification stage and seven for the abnormality detection stage. Finally, support vector machine layer is examined as a last layer of the classification in the second stage. The best average sensitivity and specificity achieved by the first stage are 95.78% and 99.45%, and 83.25% and 83.25% for the second stage, respectively. All the experiments were carried out using MURA dataset.

Download Full-text

Recalibration of deep learning models for abnormality detection in smartphone-captured chest radiograph

npj Digital Medicine ◽

10.1038/s41746-021-00393-9 ◽

2021 ◽

Vol 4 (1) ◽

Author(s):

Po-Chih Kuo ◽

Cheng Che Tsai ◽

Diego M. López ◽

Alexandros Karargyris ◽

Tom J. Pollard ◽

...

Keyword(s):

Deep Learning ◽

Operating Characteristic ◽

Characteristic Curve ◽

Model Performance ◽

X Rays ◽

Abnormality Detection ◽

Learning Models ◽

Radiological Findings ◽

Operating Characteristic Curve ◽

Uncalibrated Model

AbstractImage-based teleconsultation using smartphones has become increasingly popular. In parallel, deep learning algorithms have been developed to detect radiological findings in chest X-rays (CXRs). However, the feasibility of using smartphones to automate this process has yet to be evaluated. This study developed a recalibration method to build deep learning models to detect radiological findings on CXR photographs. Two publicly available databases (MIMIC-CXR and CheXpert) were used to build the models, and four derivative datasets containing 6453 CXR photographs were collected to evaluate model performance. After recalibration, the model achieved areas under the receiver operating characteristic curve of 0.80 (95% confidence interval: 0.78–0.82), 0.88 (0.86–0.90), 0.81 (0.79–0.84), 0.79 (0.77–0.81), 0.84 (0.80–0.88), and 0.90 (0.88–0.92), respectively, for detecting cardiomegaly, edema, consolidation, atelectasis, pneumothorax, and pleural effusion. The recalibration strategy, respectively, recovered 84.9%, 83.5%, 53.2%, 57.8%, 69.9%, and 83.0% of performance losses of the uncalibrated model. We conclude that the recalibration method can transfer models from digital CXRs to CXR photographs, which is expected to help physicians’ clinical works.

Download Full-text

White-Light Coronal Structures: Streamers and CMEs

International Astronomical Union Colloquium ◽

10.1017/s0252921100025045 ◽

1994 ◽

Vol 144 ◽

pp. 82

Author(s):

E. Hildner

Keyword(s):

Emission Line ◽

Electron Density ◽

White Light ◽

Coronal Mass Ejections ◽

X Rays ◽

Solar Cycles ◽

Temperature Function ◽

X Ray ◽

Eclipse Observations ◽

Coronal Structures

AbstractOver the last twenty years, orbiting coronagraphs have vastly increased the amount of observational material for the whitelight corona. Spanning almost two solar cycles, and augmented by ground-based K-coronameter, emission-line, and eclipse observations, these data allow us to assess,inter alia: the typical and atypical behavior of the corona; how the corona evolves on time scales from minutes to a decade; and (in some respects) the relation between photospheric, coronal, and interplanetary features. This talk will review recent results on these three topics. A remark or two will attempt to relate the whitelight corona between 1.5 and 6 R⊙to the corona seen at lower altitudes in soft X-rays (e.g., with Yohkoh). The whitelight emission depends only on integrated electron density independent of temperature, whereas the soft X-ray emission depends upon the integral of electron density squared times a temperature function. The properties of coronal mass ejections (CMEs) will be reviewed briefly and their relationships to other solar and interplanetary phenomena will be noted.

Download Full-text

Emergence of Twisted Magnetic Flux Related Sigmoidal Brightening

International Astronomical Union Colloquium ◽

10.1017/s0252921100064630 ◽

2000 ◽

Vol 179 ◽

pp. 263-264

Author(s):

K. Sundara Raman ◽

K. B. Ramesh ◽

R. Selvendran ◽

P. S. M. Aleem ◽

K. M. Hiremath

Keyword(s):

Magnetic Field ◽

Magnetic Fields ◽

Magnetic Flux ◽

Ultra Violet ◽

Active Regions ◽

Morphological Properties ◽

Energy Storage And Conversion ◽

The Sun ◽

X Rays ◽

The Magnetic Field

Extended AbstractWe have examined the morphological properties of a sigmoid associated with an SXR (soft X-ray) flare. The sigmoid is cospatial with the EUV (extreme ultra violet) images and in the optical part lies along an S-shaped Hαfilament. The photoheliogram shows flux emergence within an existingδtype sunspot which has caused the rotation of the umbrae giving rise to the sigmoidal brightening.It is now widely accepted that flares derive their energy from the magnetic fields of the active regions and coronal levels are considered to be the flare sites. But still a satisfactory understanding of the flare processes has not been achieved because of the difficulties encountered to predict and estimate the probability of flare eruptions. The convection flows and vortices below the photosphere transport and concentrate magnetic field, which subsequently appear as active regions in the photosphere (Rust & Kumar 1994 and the references therein). Successive emergence of magnetic flux, twist the field, creating flare productive magnetic shear and has been studied by many authors (Sundara Ramanet al.1998 and the references therein). Hence, it is considered that the flare is powered by the energy stored in the twisted magnetic flux tubes (Kurokawa 1996 and the references therein). Rust & Kumar (1996) named the S-shaped bright coronal loops that appear in soft X-rays as ‘Sigmoids’ and concluded that this S-shaped distortion is due to the twist developed in the magnetic field lines. These transient sigmoidal features tell a great deal about unstable coronal magnetic fields, as these regions are more likely to be eruptive (Canfieldet al.1999). As the magnetic fields of the active regions are deep rooted in the Sun, the twist developed in the subphotospheric flux tube penetrates the photosphere and extends in to the corona. Thus, it is essentially favourable for the subphotospheric twist to unwind the twist and transmit it through the photosphere to the corona. Therefore, it becomes essential to make complete observational descriptions of a flare from the magnetic field changes that are taking place in different atmospheric levels of the Sun, to pin down the energy storage and conversion process that trigger the flare phenomena.

Download Full-text