Assessing the utility of low resolution brain imaging: treatment of infant hydrocephalus

As low-field MRI technology is being disseminated into clinical settings around the world, it is important to assess the image quality required to properly diagnose and treat a given disease and evaluate the role of machine learning algorithms, such as deep learning, in the enhancement of lower qual...

Descripción completa

Guardado en:
Detalles Bibliográficos
Autores principales: Joshua R. Harper, Venkateswararao Cherukuri, Tom O’Reilly, Mingzhao Yu, Edith Mbabazi-Kabachelor, Ronald Mulando, Kevin N. Sheth, Andrew G. Webb, Benjamin C. Warf, Abhaya V. Kulkarni, Vishal Monga, Steven J. Schiff
Formato: article
Lenguaje:EN
Publicado: Elsevier 2021
Materias:
Acceso en línea:https://doaj.org/article/7bee984ac6794442a7c5a4c0b4bfc7c0
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
id oai:doaj.org-article:7bee984ac6794442a7c5a4c0b4bfc7c0
record_format dspace
spelling oai:doaj.org-article:7bee984ac6794442a7c5a4c0b4bfc7c02021-12-04T04:34:01ZAssessing the utility of low resolution brain imaging: treatment of infant hydrocephalus2213-158210.1016/j.nicl.2021.102896https://doaj.org/article/7bee984ac6794442a7c5a4c0b4bfc7c02021-01-01T00:00:00Zhttp://www.sciencedirect.com/science/article/pii/S2213158221003405https://doaj.org/toc/2213-1582As low-field MRI technology is being disseminated into clinical settings around the world, it is important to assess the image quality required to properly diagnose and treat a given disease and evaluate the role of machine learning algorithms, such as deep learning, in the enhancement of lower quality images. In this post hoc analysis of an ongoing randomized clinical trial, we assessed the diagnostic utility of reduced-quality and deep learning enhanced images for hydrocephalus treatment planning. CT images of post-infectious infant hydrocephalus were degraded in terms of spatial resolution, noise, and contrast between brain and CSF and enhanced using deep learning algorithms. Both degraded and enhanced images were presented to three experienced pediatric neurosurgeons accustomed to working in low- to middle-income countries (LMIC) for assessment of clinical utility in treatment planning for hydrocephalus. In addition, enhanced images were presented alongside their ground-truth CT counterparts in order to assess whether reconstruction errors caused by the deep learning enhancement routine were acceptable to the evaluators. Results indicate that image resolution and contrast-to-noise ratio between brain and CSF predict the likelihood of an image being characterized as useful for hydrocephalus treatment planning. Deep learning enhancement substantially increases contrast-to-noise ratio improving the apparent likelihood of the image being useful; however, deep learning enhancement introduces structural errors which create a substantial risk of misleading clinical interpretation. We find that images with lower quality than is customarily acceptable can be useful for hydrocephalus treatment planning. Moreover, low quality images may be preferable to images enhanced with deep learning, since they do not introduce the risk of misleading information which could misguide treatment decisions. These findings advocate for new standards in assessing acceptable image quality for clinical use.Joshua R. HarperVenkateswararao CherukuriTom O’ReillyMingzhao YuEdith Mbabazi-KabachelorRonald MulandoKevin N. ShethAndrew G. WebbBenjamin C. WarfAbhaya V. KulkarniVishal MongaSteven J. SchiffElsevierarticleLow field MRIImage qualityDeep learningRisk assessmentHydrocephalus treatment planningComputer applications to medicine. Medical informaticsR858-859.7Neurology. Diseases of the nervous systemRC346-429ENNeuroImage: Clinical, Vol 32, Iss , Pp 102896- (2021)
institution DOAJ
collection DOAJ
language EN
topic Low field MRI
Image quality
Deep learning
Risk assessment
Hydrocephalus treatment planning
Computer applications to medicine. Medical informatics
R858-859.7
Neurology. Diseases of the nervous system
RC346-429
spellingShingle Low field MRI
Image quality
Deep learning
Risk assessment
Hydrocephalus treatment planning
Computer applications to medicine. Medical informatics
R858-859.7
Neurology. Diseases of the nervous system
RC346-429
Joshua R. Harper
Venkateswararao Cherukuri
Tom O’Reilly
Mingzhao Yu
Edith Mbabazi-Kabachelor
Ronald Mulando
Kevin N. Sheth
Andrew G. Webb
Benjamin C. Warf
Abhaya V. Kulkarni
Vishal Monga
Steven J. Schiff
Assessing the utility of low resolution brain imaging: treatment of infant hydrocephalus
description As low-field MRI technology is being disseminated into clinical settings around the world, it is important to assess the image quality required to properly diagnose and treat a given disease and evaluate the role of machine learning algorithms, such as deep learning, in the enhancement of lower quality images. In this post hoc analysis of an ongoing randomized clinical trial, we assessed the diagnostic utility of reduced-quality and deep learning enhanced images for hydrocephalus treatment planning. CT images of post-infectious infant hydrocephalus were degraded in terms of spatial resolution, noise, and contrast between brain and CSF and enhanced using deep learning algorithms. Both degraded and enhanced images were presented to three experienced pediatric neurosurgeons accustomed to working in low- to middle-income countries (LMIC) for assessment of clinical utility in treatment planning for hydrocephalus. In addition, enhanced images were presented alongside their ground-truth CT counterparts in order to assess whether reconstruction errors caused by the deep learning enhancement routine were acceptable to the evaluators. Results indicate that image resolution and contrast-to-noise ratio between brain and CSF predict the likelihood of an image being characterized as useful for hydrocephalus treatment planning. Deep learning enhancement substantially increases contrast-to-noise ratio improving the apparent likelihood of the image being useful; however, deep learning enhancement introduces structural errors which create a substantial risk of misleading clinical interpretation. We find that images with lower quality than is customarily acceptable can be useful for hydrocephalus treatment planning. Moreover, low quality images may be preferable to images enhanced with deep learning, since they do not introduce the risk of misleading information which could misguide treatment decisions. These findings advocate for new standards in assessing acceptable image quality for clinical use.
format article
author Joshua R. Harper
Venkateswararao Cherukuri
Tom O’Reilly
Mingzhao Yu
Edith Mbabazi-Kabachelor
Ronald Mulando
Kevin N. Sheth
Andrew G. Webb
Benjamin C. Warf
Abhaya V. Kulkarni
Vishal Monga
Steven J. Schiff
author_facet Joshua R. Harper
Venkateswararao Cherukuri
Tom O’Reilly
Mingzhao Yu
Edith Mbabazi-Kabachelor
Ronald Mulando
Kevin N. Sheth
Andrew G. Webb
Benjamin C. Warf
Abhaya V. Kulkarni
Vishal Monga
Steven J. Schiff
author_sort Joshua R. Harper
title Assessing the utility of low resolution brain imaging: treatment of infant hydrocephalus
title_short Assessing the utility of low resolution brain imaging: treatment of infant hydrocephalus
title_full Assessing the utility of low resolution brain imaging: treatment of infant hydrocephalus
title_fullStr Assessing the utility of low resolution brain imaging: treatment of infant hydrocephalus
title_full_unstemmed Assessing the utility of low resolution brain imaging: treatment of infant hydrocephalus
title_sort assessing the utility of low resolution brain imaging: treatment of infant hydrocephalus
publisher Elsevier
publishDate 2021
url https://doaj.org/article/7bee984ac6794442a7c5a4c0b4bfc7c0
work_keys_str_mv AT joshuarharper assessingtheutilityoflowresolutionbrainimagingtreatmentofinfanthydrocephalus
AT venkateswararaocherukuri assessingtheutilityoflowresolutionbrainimagingtreatmentofinfanthydrocephalus
AT tomoreilly assessingtheutilityoflowresolutionbrainimagingtreatmentofinfanthydrocephalus
AT mingzhaoyu assessingtheutilityoflowresolutionbrainimagingtreatmentofinfanthydrocephalus
AT edithmbabazikabachelor assessingtheutilityoflowresolutionbrainimagingtreatmentofinfanthydrocephalus
AT ronaldmulando assessingtheutilityoflowresolutionbrainimagingtreatmentofinfanthydrocephalus
AT kevinnsheth assessingtheutilityoflowresolutionbrainimagingtreatmentofinfanthydrocephalus
AT andrewgwebb assessingtheutilityoflowresolutionbrainimagingtreatmentofinfanthydrocephalus
AT benjamincwarf assessingtheutilityoflowresolutionbrainimagingtreatmentofinfanthydrocephalus
AT abhayavkulkarni assessingtheutilityoflowresolutionbrainimagingtreatmentofinfanthydrocephalus
AT vishalmonga assessingtheutilityoflowresolutionbrainimagingtreatmentofinfanthydrocephalus
AT stevenjschiff assessingtheutilityoflowresolutionbrainimagingtreatmentofinfanthydrocephalus
_version_ 1718372977251188736