Spine-GFlow: A hybrid learning framework for robust multi-tissue segmentation in lumbar MRI without manual annotation.

Most learning-based magnetic resonance image (MRI) segmentation methods rely on the manual annotation to provide supervision, which is extremely tedious, especially when multiple anatomical structures are required. In this work, we aim to develop a hybrid framework named Spine-GFlow that combines the image features learned by a CNN model and anatomical priors for multi-tissue segmentation in a sagittal lumbar MRI. Our framework does not require any manual annotation and is robust against image feature variation caused by different image settings and/or underlying pathology. Our contributions include: 1) a rule-based method that automatically generates the weak annotation (initial seed area), 2) a novel proposal generation method that integrates the multi-scale image features and anatomical prior, 3) a comprehensive loss for CNN training that optimizes the pixel classification and feature distribution simultaneously. Our Spine-GFlow has been validated on 2 independent datasets: HKDDC (containing images obtained from 3 different machines) and IVDM3Seg. The segmentation results of vertebral bodies (VB), intervertebral discs (IVD), and spinal canal (SC) are evaluated quantitatively using intersection over union (IoU) and the Dice coefficient. Results show that our method, without requiring manual annotation, has achieved a segmentation performance comparable to a model trained with full supervision (mean Dice 0.914 vs 0.916).

An artificial intelligence powered platform for auto-analyses of spine alignment irrespective of image quality with prospective validation.

BACKGROUND: Assessment of spine alignment is crucial in the management of scoliosis, but current auto-analysis of spine alignment suffers from low accuracy. We aim to develop and validate a hybrid model named SpineHRNet+, which integrates artificial intelligence (AI) and rule-based methods to improve auto-alignment reliability and interpretability. METHODS: From December 2019 to November 2020, 1,542 consecutive patients with scoliosis attending two local scoliosis clinics (The Duchess of Kent Children's Hospital at Sandy Bay in Hong Kong; Queen Mary Hospital in Pok Fu Lam on Hong Kong Island) were recruited. The biplanar radiographs of each patient were collected with our medical machine EOS™. The collected radiographs were recaptured using smartphones or screenshots, with deidentified images securely stored. Manually labelled landmarks and alignment parameters by a spine surgeon were considered as ground truth (GT). The data were split 8:2 to train and internally test SpineHRNet+, respectively. This was followed by a prospective validation on another 337 patients. Quantitative analyses of landmark predictions were conducted, and reliabilities of auto-alignment were assessed using linear regression and Bland-Altman plots. Deformity severity and sagittal abnormality classifications were evaluated by confusion matrices. FINDINGS: SpineHRNet+ achieved accurate landmark detection with mean Euclidean distance errors of 2·78 and 5·52 pixels on posteroanterior and lateral radiographs, respectively. The mean angle errors between predictions and GT were 3·18° and 6·32° coronally and sagittally. All predicted alignments were strongly correlated with GT (p < 0·001, R2 > 0·97), with minimal overall difference visualised via Bland-Altman plots. For curve detections, 95·7% sensitivity and 88·1% specificity was achieved, and for severity classification, 88·6-90·8% sensitivity was obtained. For sagittal abnormalities, greater than 85·2-88·9% specificity and sensitivity were achieved. INTERPRETATION: The auto-analysis provided by SpineHRNet+ was reliable and continuous and it might offer the potential to assist clinical work and facilitate large-scale clinical studies. FUNDING: RGC Research Impact Fund (R5017-18F), Innovation and Technology Fund (ITS/404/18), and the AOSpine East Asia Fund (AOSEA(R)2019-06).

Global retinoblastoma survival and globe preservation: a systematic review and meta-analysis of associations with socioeconomic and health-care factors.

BACKGROUND: Despite advancements in globe-preserving treatments, improvements in retinoblastoma outcomes are inconsistent across income levels and geographical locations. We aimed to investigate trends in global retinoblastoma survival and globe preservation during the past 40 years. We also examined associated socioeconomic and health-care factors and global survival disparity. METHODS: We did a systematic review and meta-analysis by screening articles in any language in nine databases (PubMed, Embase, ScienceDirect, Web of Science, OpenGrey, Global Burden of Disease, Global Health Data Exchange, Global Index Medicus, and International Agency for the Prevention of Blindness) published between Jan 1, 1981, and Oct 8, 2021. We screened for articles that described retinoblastoma overall survival or globe salvage, or both. All reported studies were subsequently stratified into four periods: 1980-89, 1990-99, 2000-09, and 2010-20. Indicators on socioeconomic and health-care factors were extracted from the World Bank and WHO. Ophthalmology-related indicators were further parsed from the International Agency for the Prevention of Blindness. Between-study heterogeneities by income level were assessed by mixed-effect meta-analysis. Associations of retinoblastoma outcome with socioeconomic and health-care factors and factors for survival prediction were investigated by multivariable linear regressions. This study is registered with PROSPERO, number CRD42020221556. FINDINGS: Our search identified 14 621 articles, of which 314 studies were included for analysis after screening, including 38 130 patients from 80 regions globally presenting during 1980-2020. 255 articles were entered for time-trend meta-analysis, covering 29 106 patients from 73 countries. Both overall survival (from 79% [95% CI 74-84] to 88% [83-93]; p=0·017) and globe salvage rate (from 22% [14-32] to 44% [36-52]; p=0·0003) improved significantly over the four decades. Wide disparities were observed between higher-income and lower-income countries. Overall survival, globe salvage, and globe salvage for advanced intraocular disease correlated positively with income level. Higher overall survival was associated with lower Gini index (p=0·0001) and with populations that had smaller percentages living in rural areas (p=0·0005). Higher globe salvage was associated with better health-care financing and accessibility (p=0·030). Overall survival (p=0·0024) and globe salvage (p=0·022) were both associated positively with education level. Survival gaps were observed in sub-Saharan Africa and southeast and southwest Asia. INTERPRETATION: Retinoblastoma treatment outcomes have improved globally over the past four decades but large disparities persist between higher-income and lower-income countries, with some areas having major survival gaps. Targeted health-care policy making with increased health-care financing and accessibility are needed in low-income and lower-middle-income countries to improve retinoblastoma outcomes worldwide. FUNDING: Health and Medical Research Fund (Hong Kong) and Children Cancer's Foundation (Hong Kong).