Using DICOM Metadata for Radiological Image Series Categorization: a Feasibility Study on Large Clinical Brain MRI Datasets

Gauriau, Romane; Bridge, Christopher P.; Chen, Lina; Kitamura, Felipe; Tenenholtz, Neil A.; Kirsch, John E.; Andriole, Katherine P.; Michalski, Mark; Bizzo, Bernardo C.

doi:10.1007/s10278-019-00308-x

Cited by 31 publications

(25 citation statements)

References 27 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Brain MR-DWI and Apparent Diffusion Coe cient (ADC) sequences were considered ground truth for the presence or absence of acute infarction; axial DWI "b=1000" and ADC series with slice thickness ≥5mm were selected using a brain MRI series selection algorithm [22]. All images were reviewed by a trained radiologist (JKC, DC, BB, JP, AP, IS, JC) to ensure correct classi cation.…”

Section: Methodsmentioning

confidence: 99%

Head CT Deep Learning Model for Early Stroke Identification Outperforms Human Experts

Bizzo¹,

Gauriau²,

Comeau³

et al. 2021

Preprint

Self Cite

View full text Add to dashboard Cite

Non-contrast head CT (NCCT) is extremely insensitive for early (< 3-6hrs) acute infarct identification. We developed a deep learning model that detects and delineates early acute infarcts on NCCT, using diffusion MRI as ground truth (3,566 NCCT/MRI training pairs). The model substantially outperformed 3 expert neuroradiologists on a test set of 150 CT scans (sensitivity 96% model versus 61–66% experts); infarct volume estimates strongly correlated with those of diffusion MRI (r2 > 0.98).

show abstract

Section: Methodsmentioning

confidence: 99%

Head CT Deep Learning Model for Early Stroke Identification Outperforms Human Experts

Bizzo¹,

Gauriau²,

Comeau³

et al. 2021

Preprint

Self Cite

View full text Add to dashboard Cite

show abstract

“…The dataset breakdown allowed us to cover greater than 90% of abnormalities (see Supplemental Material). Axial FLAIR sequences were selected automatically (17). Tables 1 and 2show datasets splits, acquisition, patient, and annotations details, highlighting dataset heterogeneity.…”

Section: Training Validation and Test Datasetsmentioning

confidence: 99%

A Deep Learning–based Model for Detecting Abnormalities on Brain MR Images for Triaging: Preliminary Results from a Multisite Experience

Gauriau¹,

Bizzo²,

Kitamura³

et al. 2021

Radiology: Artificial Intelligence

Self Cite

View full text Add to dashboard Cite

To develop a deep learning model for detecting brain abnormalities on MRI. Materials and Methods:In this retrospective study, a deep learning approach using T2-weighted fluid attenuated inversion recovery (FLAIR) images was developed to classify brain MRI as "likely normal" or "likely abnormal." A convolutional neural network model was trained on a large heterogeneous dataset collected from two different continents and covering a broad panel of pathologies including neoplasms, hemorrhage, infarcts, and others. Three datasets were used. Dataset A consisted of 2839 patients, Dataset B consisted of 6442 patients, and Dataset C consisted of 1489 patients and was only used for testing. Dataset A and B were split into training, validation and test sets. A total of three models were trained: Model A (using only Dataset A), Model B (using only Dataset B), and Model A+B (using training datasets from A and B). All three models were tested on subsets from Dataset A, Dataset B and Dataset C separately. The evaluation was performed using annotations based on the images as well as labels based on the radiologic reports.

show abstract

“…Current practice also suggests using the DICOM attribute Image Orientation Patient (0020,0037). The attribute values give the direction cosines of the first row and first column of an image with respect to the patient which could be further used to compute the orientation plane of the scan by computing the main direction of the normal to the slices [2]. But even in the aforementioned method missing slices or extra slices, unidentified attributes and unexpected errors may cause difficulty in detecting the orientation plane of the brain MR scan.There is very little work done to detect the orientation plane of brain MR scans especially using pixel data and deep learning.…”

Section: Background Informationmentioning

confidence: 99%

“…The current works in this area suggest methods relying on the metadata of the DICOM files but the real question is how reliable is the metadata? The values of various attributes in the DICOM metadata are unreliable and inconsistent [2,3,4]. Automating the classification of the orientation plane of the brain MR scans thus becomes challenging.…”

Section: Introductionmentioning

confidence: 99%

“…While many entities are working on the development of significant deep learning models for the diagnosis of brain-related diseases, identifying precise images needed for model training and inference tasks is limited due to variation in DICOM fields which use free text to define things like series description, sequence and orientation [1]. Detecting the orientation of brain MR scans (Axial/Sagittal/Coronal) remains a challenge due to these variations caused by linguistic barriers, human errors and de-identification -essentially rendering the tags unreliable [2,3,4]. In this work, we propose a deep learning model that identifies the orientation of brain MR scans with near perfect accuracy.…”

mentioning

confidence: 99%

See 1 more Smart Citation

Detecting orientation of Brain MR scans using deep learning

Singhal¹,

Gupta

Stein³

et al. 2021

Preprint

View full text Add to dashboard Cite

There has been a steady escalation in the impact of Artificial Intelligence (AI) on Healthcare along with an increasing amount of progress being made in this field. While many entities are working on the development of significant deep learning models for the diagnosis of brain-related diseases, identifying precise images needed for model training and inference tasks is limited due to variation in DICOM fields which use free text to define things like series description, sequence and orientation \cite{willemink_koszek_hardell_wu_fleischmann_harvey_folio_summers_rubin_lungren_2020}. Detecting the orientation of brain MR scans (Axial/Sagittal/Coronal) remains a challenge due to these variations caused by linguistic barriers, human errors and de-identification - essentially rendering the tags unreliable. In this work, we propose a deep learning model that identifies the orientation of brain MR scans with near perfect accuracy.

show abstract

Using DICOM Metadata for Radiological Image Series Categorization: a Feasibility Study on Large Clinical Brain MRI Datasets

Cited by 31 publications

References 27 publications

Head CT Deep Learning Model for Early Stroke Identification Outperforms Human Experts

Head CT Deep Learning Model for Early Stroke Identification Outperforms Human Experts

A Deep Learning–based Model for Detecting Abnormalities on Brain MR Images for Triaging: Preliminary Results from a Multisite Experience

Detecting orientation of Brain MR scans using deep learning

Contact Info

Product

Resources

About