-
Self-supervised Skull Reconstruction in Brain CT Images with Decompressive Craniectomy
Authors:
Franco Matzkin,
Virginia Newcombe,
Susan Stevenson,
Aneesh Khetani,
Tom Newman,
Richard Digby,
Andrew Stevens,
Ben Glocker,
Enzo Ferrante
Abstract:
Decompressive craniectomy (DC) is a common surgical procedure consisting of the removal of a portion of the skull that is performed after incidents such as stroke, traumatic brain injury (TBI) or other events that could result in acute subdural hemorrhage and/or increasing intracranial pressure. In these cases, CT scans are obtained to diagnose and assess injuries, or guide a certain therapy and i…
▽ More
Decompressive craniectomy (DC) is a common surgical procedure consisting of the removal of a portion of the skull that is performed after incidents such as stroke, traumatic brain injury (TBI) or other events that could result in acute subdural hemorrhage and/or increasing intracranial pressure. In these cases, CT scans are obtained to diagnose and assess injuries, or guide a certain therapy and intervention.
We propose a deep learning based method to reconstruct the skull defect removed during DC performed after TBI from post-operative CT images. This reconstruction is useful in multiple scenarios, e.g. to support the creation of cranioplasty plates, accurate measurements of bone flap volume and total intracranial volume, important for studies that aim to relate later atrophy to patient outcome. We propose and compare alternative self-supervised methods where an encoder-decoder convolutional neural network (CNN) estimates the missing bone flap on post-operative CTs. The self-supervised learning strategy only requires images with complete skulls and avoids the need for annotated DC images. For evaluation, we employ real and simulated images with DC, comparing the results with other state-of-the-art approaches. The experiments show that the proposed model outperforms current manual methods, enabling reconstruction even in highly challenging cases where big skull defects have been removed during surgery.
△ Less
Submitted 10 July, 2020; v1 submitted 7 July, 2020;
originally announced July 2020.
-
The star formation histories of dwarf galaxies in Local Group cosmological simulations
Authors:
Ruth Digby,
Julio F. Navarro,
Azadeh Fattahi,
Christine M. Simpson,
Kyle A. Oman,
Facundo A. Gomez,
Carlos S. Frenk,
Robert J. J. Grand,
Ruediger Pakmor
Abstract:
We use the APOSTLE and Auriga cosmological simulations to study the star formation histories (SFHs) of field and satellite dwarf galaxies. Despite sizeable galaxy-to-galaxy scatter, the SFHs of APOSTLE and Auriga dwarfs exhibit robust average trends with galaxy stellar mass: faint field dwarfs ($10^5<M_{\rm star}/M_\odot<10^{6.5}$) have, on average, steadily declining SFHs, whereas brighter dwarfs…
▽ More
We use the APOSTLE and Auriga cosmological simulations to study the star formation histories (SFHs) of field and satellite dwarf galaxies. Despite sizeable galaxy-to-galaxy scatter, the SFHs of APOSTLE and Auriga dwarfs exhibit robust average trends with galaxy stellar mass: faint field dwarfs ($10^5<M_{\rm star}/M_\odot<10^{6.5}$) have, on average, steadily declining SFHs, whereas brighter dwarfs ($10^{7.5}<M_{\rm star}/M_\odot<10^{9}$) show the opposite trend. Intermediate-mass dwarfs have roughly constant SFHs. Satellites exhibit similar average trends, but with substantially suppressed star formation in the most recent $\sim 5$ Gyr, likely as a result of gas loss due to tidal and ram-pressure strip** after entering the haloes of their primaries. These simple mass and environmental trends are in good agreement with the derived SFHs of Local Group (LG) dwarfs whose photometry reaches the oldest main sequence turnoff. SFHs of galaxies with less deep data show deviations from these trends, but this may be explained, at least in part, by the large galaxy-to-galaxy scatter, the limited sample size, and the large uncertainties of the inferred SFHs. Confirming the predicted mass and environmental trends will require deeper photometric data than currently available, especially for isolated dwarfs.
△ Less
Submitted 12 March, 2019; v1 submitted 13 December, 2018;
originally announced December 2018.
-
The Peculiar Globular Cluster Palomar 1 and Persistence in the SDSS-APOGEE Database
Authors:
Farbod Jahandar,
Kim A. Venn,
Matthew D. Shetrone,
Mike Irwin,
Jo Bovy,
Charli M. Sakari,
Collin L. Kielty,
Ruth A. R. Digby,
Peter M. Frinchaboy
Abstract:
The SDSS-III APOGEE DR12 is a unique resource to search for stars beyond the tidal radii of star clusters. We have examined the APOGEE DR12 database for new candidates of the young star cluster Palomar 1, a system with previously reported tidal tails (Niederste-Ostholt et al. 2010). The APOGEE ASPCAP database includes spectra and stellar parameters for two known members of Pal 1 (Stars I and II),…
▽ More
The SDSS-III APOGEE DR12 is a unique resource to search for stars beyond the tidal radii of star clusters. We have examined the APOGEE DR12 database for new candidates of the young star cluster Palomar 1, a system with previously reported tidal tails (Niederste-Ostholt et al. 2010). The APOGEE ASPCAP database includes spectra and stellar parameters for two known members of Pal 1 (Stars I and II), however these do not agree with the stellar parameters determined from optical spectra by Sakari et al. (2011). We find that the APOGEE analysis of these two stars is strongly affected by the known persistence problem (Majewski et al. 2015; Nidever et al. 2015). By re-examining the individual visits, and removing the blue (and sometimes green) APOGEE detector spectra affected by persistence, then we find excellent agreement in a re-analysis of the combined spectra. These methods are applied to another five stars in the APOGEE field with similar radial velocities and metallicities as those of Pal 1. Only one of these new candidates, Star F, may be a member located in the tidal tail based on its heliocentric radial velocity, metallicity, and chemistry. The other four candidates are not well aligned with the tidal tails, and comparison to the Besancon model (Robin et al. 2003) suggests that they are more likely to be non-members, i.e. part of the Galactic halo. This APOGEE field could be re-examined for other new candidates if the persistence problem can be removed from the APOGEE spectral database.
△ Less
Submitted 27 June, 2017;
originally announced June 2017.
