1
|
Roberts EJ, Chavez T, Hexemer A, Zwart PH. DLSIA: Deep Learning for Scientific Image Analysis. J Appl Crystallogr 2024; 57:392-402. [PMID: 38596727 PMCID: PMC11001410 DOI: 10.1107/s1600576724001390] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 09/15/2023] [Accepted: 02/12/2024] [Indexed: 04/11/2024] Open
Abstract
DLSIA (Deep Learning for Scientific Image Analysis) is a Python-based machine learning library that empowers scientists and researchers across diverse scientific domains with a range of customizable convolutional neural network (CNN) architectures for a wide variety of tasks in image analysis to be used in downstream data processing. DLSIA features easy-to-use architectures, such as autoencoders, tunable U-Nets and parameter-lean mixed-scale dense networks (MSDNets). Additionally, this article introduces sparse mixed-scale networks (SMSNets), generated using random graphs, sparse connections and dilated convolutions connecting different length scales. For verification, several DLSIA-instantiated networks and training scripts are employed in multiple applications, including inpainting for X-ray scattering data using U-Nets and MSDNets, segmenting 3D fibers in X-ray tomographic reconstructions of concrete using an ensemble of SMSNets, and leveraging autoencoder latent spaces for data compression and clustering. As experimental data continue to grow in scale and complexity, DLSIA provides accessible CNN construction and abstracts CNN complexities, allowing scientists to tailor their machine learning approaches, accelerate discoveries, foster interdisciplinary collaboration and advance research in scientific image analysis.
Collapse
Affiliation(s)
- Eric J. Roberts
- Center for Advanced Mathematics for Energy Research Applications, Lawrence Berkeley National Laboratory, Berkeley, CA 94720, USA
- Molecular Biophysics and Integrated Bioimaging Division, Lawrence Berkeley National Laboratory, Berkeley, CA 94720, USA
| | - Tanny Chavez
- Advanced Light Source, Lawrence Berkeley National Laboratory, Berkeley, CA 94720, USA
| | - Alexander Hexemer
- Center for Advanced Mathematics for Energy Research Applications, Lawrence Berkeley National Laboratory, Berkeley, CA 94720, USA
- Advanced Light Source, Lawrence Berkeley National Laboratory, Berkeley, CA 94720, USA
| | - Petrus H. Zwart
- Center for Advanced Mathematics for Energy Research Applications, Lawrence Berkeley National Laboratory, Berkeley, CA 94720, USA
- Molecular Biophysics and Integrated Bioimaging Division, Lawrence Berkeley National Laboratory, Berkeley, CA 94720, USA
- Berkeley Synchrotron Infrared Structural Biology Program, Lawrence Berkeley National Laboratory, Berkeley, CA 94720, USA
| |
Collapse
|
2
|
Besnard C, Marie A, Sasidharan S, Buček P, Walker JM, Parker JE, Spink MC, Harper RA, Marathe S, Wanelik K, Moxham TE, Salvati E, Ignatyev K, Kłosowski MM, Shelton RM, Landini G, Korsunsky AM. Multi-resolution Correlative Ultrastructural and Chemical Analysis of Carious Enamel by Scanning Microscopy and Tomographic Imaging. ACS APPLIED MATERIALS & INTERFACES 2023; 15:37259-37273. [PMID: 37524079 PMCID: PMC10416148 DOI: 10.1021/acsami.3c08031] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Grants] [Track Full Text] [Subscribe] [Scholar Register] [Received: 06/05/2023] [Accepted: 07/17/2023] [Indexed: 08/02/2023]
Abstract
Caries, a major global disease associated with dental enamel demineralization, remains insufficiently understood to devise effective prevention or minimally invasive treatment. Understanding the ultrastructural changes in enamel is hampered by a lack of nanoscale characterization of the chemical spatial distributions within the dental tissue. This leads to the requirement to develop techniques based on various characterization methods. The purpose of the present study is to demonstrate the strength of analytic methods using a correlative technique on a single sample of human dental enamel as a specific case study to test the accuracy of techniques to compare regions in enamel. The science of the different techniques is integrated to genuinely study the enamel. The hierarchical structures within carious tissue were mapped using the combination of focused ion beam scanning electron microscopy with synchrotron X-ray tomography. The chemical changes were studied using scanning X-ray fluorescence (XRF) and X-ray wide-angle and small-angle scattering using a beam size below 80 nm for ångström and nanometer length scales. The analysis of XRF intensity gradients revealed subtle variations of Ca intensity in carious samples in comparison with those of normal mature enamel. In addition, the pathways for enamel rod demineralization were studied using X-ray ptychography. The results show the chemical and structural modification in carious enamel with differing locations. These results reinforce the need for multi-modal approaches to nanoscale analysis in complex hierarchically structured materials to interpret the changes of materials. The approach establishes a meticulous correlative characterization platform for the analysis of biomineralized tissues at the nanoscale, which adds confidence in the interpretation of the results and time-saving imaging techniques. The protocol demonstrated here using the dental tissue sample can be applied to other samples for statistical study and the investigation of nanoscale structural changes. The information gathered from the combination of methods could not be obtained with traditional individual techniques.
Collapse
Affiliation(s)
- Cyril Besnard
- MBLEM,
Department of Engineering Science, University
of Oxford, Parks Road, Oxford, Oxfordshire OX1
3PJ, U.K.
| | - Ali Marie
- MBLEM,
Department of Engineering Science, University
of Oxford, Parks Road, Oxford, Oxfordshire OX1
3PJ, U.K.
| | - Sisini Sasidharan
- MBLEM,
Department of Engineering Science, University
of Oxford, Parks Road, Oxford, Oxfordshire OX1
3PJ, U.K.
| | - Petr Buček
- TESCAN-UK
Ltd., Wellbrook Court, Girton, Cambridge CB3 0NA, U.K.
| | | | - Julia E. Parker
- Diamond
Light Source Ltd., Didcot, Oxfordshire OX11 0DE, U.K.
| | | | - Robert A. Harper
- School
of Dentistry, University of Birmingham, 5 Mill Pool Way, Edgbaston, Birmingham, West Midlands B5 7EG, U.K.
| | | | - Kaz Wanelik
- Diamond
Light Source Ltd., Didcot, Oxfordshire OX11 0DE, U.K.
| | - Thomas E.J. Moxham
- MBLEM,
Department of Engineering Science, University
of Oxford, Parks Road, Oxford, Oxfordshire OX1
3PJ, U.K.
- Diamond
Light Source Ltd., Didcot, Oxfordshire OX11 0DE, U.K.
| | - Enrico Salvati
- MBLEM,
Department of Engineering Science, University
of Oxford, Parks Road, Oxford, Oxfordshire OX1
3PJ, U.K.
| | | | | | - Richard M. Shelton
- School
of Dentistry, University of Birmingham, 5 Mill Pool Way, Edgbaston, Birmingham, West Midlands B5 7EG, U.K.
| | - Gabriel Landini
- School
of Dentistry, University of Birmingham, 5 Mill Pool Way, Edgbaston, Birmingham, West Midlands B5 7EG, U.K.
| | - Alexander M. Korsunsky
- MBLEM,
Department of Engineering Science, University
of Oxford, Parks Road, Oxford, Oxfordshire OX1
3PJ, U.K.
| |
Collapse
|
3
|
Besnard C, Marie A, Sasidharan S, Harper RA, Shelton RM, Landini G, Korsunsky AM. Synchrotron X-ray Studies of the Structural and Functional Hierarchies in Mineralised Human Dental Enamel: A State-of-the-Art Review. Dent J (Basel) 2023; 11:98. [PMID: 37185477 PMCID: PMC10137518 DOI: 10.3390/dj11040098] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Grants] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 01/04/2023] [Revised: 03/19/2023] [Accepted: 03/28/2023] [Indexed: 05/17/2023] Open
Abstract
Hard dental tissues possess a complex hierarchical structure that is particularly evident in enamel, the most mineralised substance in the human body. Its complex and interlinked organisation at the Ångstrom (crystal lattice), nano-, micro-, and macro-scales is the result of evolutionary optimisation for mechanical and functional performance: hardness and stiffness, fracture toughness, thermal, and chemical resistance. Understanding the physical-chemical-structural relationships at each scale requires the application of appropriately sensitive and resolving probes. Synchrotron X-ray techniques offer the possibility to progress significantly beyond the capabilities of conventional laboratory instruments, i.e., X-ray diffractometers, and electron and atomic force microscopes. The last few decades have witnessed the accumulation of results obtained from X-ray scattering (diffraction), spectroscopy (including polarisation analysis), and imaging (including ptychography and tomography). The current article presents a multi-disciplinary review of nearly 40 years of discoveries and advancements, primarily pertaining to the study of enamel and its demineralisation (caries), but also linked to the investigations of other mineralised tissues such as dentine, bone, etc. The modelling approaches informed by these observations are also overviewed. The strategic aim of the present review was to identify and evaluate prospective avenues for analysing dental tissues and developing treatments and prophylaxis for improved dental health.
Collapse
Affiliation(s)
- Cyril Besnard
- MBLEM, Department of Engineering Science, University of Oxford, Parks Road, Oxford OX1 3PJ, Oxfordshire, UK
| | - Ali Marie
- MBLEM, Department of Engineering Science, University of Oxford, Parks Road, Oxford OX1 3PJ, Oxfordshire, UK
| | - Sisini Sasidharan
- MBLEM, Department of Engineering Science, University of Oxford, Parks Road, Oxford OX1 3PJ, Oxfordshire, UK
| | - Robert A. Harper
- School of Dentistry, University of Birmingham, 5 Mill Pool Way, Edgbaston, Birmingham B5 7EG, West Midlands, UK
| | - Richard M. Shelton
- School of Dentistry, University of Birmingham, 5 Mill Pool Way, Edgbaston, Birmingham B5 7EG, West Midlands, UK
| | - Gabriel Landini
- School of Dentistry, University of Birmingham, 5 Mill Pool Way, Edgbaston, Birmingham B5 7EG, West Midlands, UK
| | - Alexander M. Korsunsky
- MBLEM, Department of Engineering Science, University of Oxford, Parks Road, Oxford OX1 3PJ, Oxfordshire, UK
| |
Collapse
|
4
|
Gan SC, Fok ASL, Sedky RA, Sukumaran P, Chew HP. Sorptivity of water in enamel for categorizing caries lesions. Dent Mater 2020; 36:1379-1387. [PMID: 32907752 DOI: 10.1016/j.dental.2020.08.003] [Citation(s) in RCA: 1] [Impact Index Per Article: 0.3] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 07/21/2020] [Revised: 07/29/2020] [Accepted: 08/15/2020] [Indexed: 11/28/2022]
Abstract
OBJECTIVE The aim of this research was to determine the association between sorptivity of water and the state of mineralization in carious enamel of different stages of severity. METHODS As a preliminary work, water droplets of 1.5 μL were placed on the surfaces of hydroxyapatite (HA) discs of different densities. The water droplet profile was dynamically recorded every second over a period of 10 s using a contact angle meter to determine the relationship between sorptivity and density. To measure and calculate sorptivity on enamel surfaces, varnish was painted on the labial surface of 96 extracted caries-free human teeth, leaving two 1.4 ± 0.1 mm diameter circular exposed test sites. The specimens were randomly divided into 6 groups (n = 16) and subjected to 0(G0), 7(G7), 14(G14), 21(G21), 28(G28) and 35(G35) days of pH cycling, respectively. A 0.7 μL water droplet was placed on each exposed site and Optical Coherence Tomography was used to measure its height every 10 seconds for 2 min. Sorptivity was computed by considering sorption equations and Washburn's analysis of capillary kinetics and correction for evaporation was also performed. Micro-Computed Tomography scans of the specimens were obtained and delta Z (ΔZ) is the parameter used to measure mineral loss. ΔZ at 10 μm (ΔZ10) and 50 μm (ΔZ50) from the surface were calculated. One-way ANOVA and Post-hoc Tukey tests were used to compare sorptivity between groups and bivariate correlations were used to analyze the association between sorptivity and ΔZ. RESULTS Sorptivity was found to be inversely and linearly correlated with HA density with R2 value of 0.95. With enamel, there is a general trend of increase in mean sorptivity from G0 to G35, except for a decrease in G21. The same trends were observed for both ΔZ10 and ΔZ50. The decrease in sorptivity in G21 coincided with the presence of a surface hypermineralized layer in G21 samples. Post-hoc Tukey showed significant differences in mean sorptivity between G0 and G14, G0 and G21 as well as G14 and G21. Post-hoc Dunnett's T3 showed significant differences for ΔZ10 between G0 and G14 as well as G14 and G21. Significant correlation between mean sorptivity and ΔZ10 was detected with Pearson correlation coefficient of 0.461. For ΔZ50, post-hoc Tukey showed significant differences between G0 and G14 but no significant difference was detected between G14 and G21. No correlations were detected between mean sorptivity and ΔZ50. SIGNIFICANCE Sorptivity was found to be inversely and linearly correlated with HA density with R2 value of 0.95. With enamel, there is a general trend of increase in mean sorptivity from G0 to G35, except for a decrease in G21. The same trends were observed for both ΔZ10 and ΔZ50. The decrease in sorptivity in G21 coincided with the presence of a surface hypermineralized layer in G21 samples.
Collapse
Affiliation(s)
| | - Alex Siu Lun Fok
- Minnesota Dental Research Center for Biomaterials and Biomechanics, School of Dentistry, University of Minnesota, USA
| | - Rana A Sedky
- Faculty of Dentistry, Ain-Shams University, Cairo, Egypt
| | | | - Hooi Pin Chew
- Minnesota Dental Research Center for Biomaterials and Biomechanics, School of Dentistry, University of Minnesota, USA.
| |
Collapse
|
5
|
Phase-Contrast and Dark-Field Imaging. J Imaging 2018. [DOI: 10.3390/jimaging4100113] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 11/17/2022] Open
Abstract
Very early, in 1896, Wilhelm Conrad Röntgen, the founding father of X-rays, attempted to measure diffraction and refraction by this new kind of radiation, in vain. Only 70 years later, these effects were measured by Ulrich Bonse and Michael Hart who used them to make full-field images of biological specimen, coining the term phase-contrast imaging. Yet, another 30 years passed until the Talbot effect was rediscovered for X-radiation, giving rise to a micrograting based interferometer, replacing the Bonse–Hart interferometer, which relied on a set of four Laue-crystals for beam splitting and interference. By merging the Lau-interferometer with this Talbot-interferometer, another ten years later, measuring X-ray refraction and X-ray scattering full-field and in cm-sized objects (as Röntgen had attempted 110 years earlier) became feasible in every X-ray laboratory around the world. Today, now that another twelve years have passed and we are approaching the 125th jubilee of Röntgen’s discovery, neither Laue-crystals nor microgratings are a necessity for sensing refraction and scattering by X-rays. Cardboard, steel wool, and sandpaper are sufficient for extracting these contrasts from transmission images, using the latest image reconstruction algorithms. This advancement and the ever rising number of applications for phase-contrast and dark-field imaging prove to what degree our understanding of imaging physics as well as signal processing have advanced since the advent of X-ray physics, in particular during the past two decades. The discovery of the electron, as well as the development of electron imaging technology, has accompanied X-ray physics closely along its path, both modalities exploring the applications of new dark-field contrast mechanisms these days. Materials science, life science, archeology, non-destructive testing, and medicine are the key faculties which have already integrated these new imaging devices, using their contrast mechanisms in full. This special issue “Phase-Contrast and Dark-field Imaging” gives us a broad yet very to-the-point glimpse of research and development which are currently taking place in this very active field. We find reviews, applications reports, and methodological papers of very high quality from various groups, most of which operate X-ray scanners which comprise these new imaging modalities.
Collapse
|