Background It is unclear whether newborns with mild TSH elevation (mTSHe) are at risk of neurocognitive impairment. We assessed whether mTSHe at birth persists during childhood and compared neurocognitive functioning to siblings. Methods This study encompassed children born in the Auckland region with a newborn screen TSH level 8–14 mIU/L blood, aged 6.9–12.6 years at assessment, and their siblings. Thyroid function tests (serum TSH and FT4) and neurocognitive assessments were performed, including intelligent quotient (IQ) via Wechsler Intelligence Scale for Children IV. Results Ninety-six mTSHe subjects were studied, including 67 children recruited with 75 sibling controls. Mean mTSHe newborn TSH level was 10.1 mIU/L (range 8–14 mIU/L, blood) and 2.4 mIU/L at assessment (range 0.8–7.0 mIU/L, serum). Although higher newborn TSH levels in the mTSHe group correlated with lower full-scale IQ scores (r=0.25; p=0.040), they were not associated with the magnitude of the IQ difference within sibling pairs (p=0.56). Cognitive scores were similar for mTSHe and controls (full-scale IQ 107 vs 109; p=0.36), with a minor isolated difference in motor co-ordination scores. Conclusions Our data do not demonstrate long-term negative effects of neonatal mild TSH elevation. TSH elevation below screen threshold appears largely transient and mid-childhood neurocognitive performance of these children was similar to their siblings. We propose that associations between neonatal mild TSH elevation and IQ are due to familial confounders. We caution against the practice of reducing screening CH cutoffs to levels where the diagnosis may not offer long-term benefit for those detected.

ObjectivePatients with thalassemia major do require lifetime blood transfusions that eventually result in iron accumulation in different organs. We described the usefulness of using magnetic resonance imaging (MRI) T2*imaging values for the evaluation of pancreatic iron load in these patients, and we correlated it with MRI T2* haemosiderosis of the myocardium and liver that has been recognized as a non-invasive assessment of iron overload among patients with thalassemia major.Materials and methodsWe conducted a cross-sectional study on 39 patients with thalassemia major in one of the tertiary university hospitals for a 1-year period. Demographic data were collected from the patient's history. MRI T2* of the pancreas, liver, and heart were executed on all patients in the same setting. Objective values of iron overload in these organs were obtained using the MRI post-processing software from online software.ResultsA total of 32 (82.1%) patients had pancreatic iron overload including 2 patients (5.1%) with severe iron overload and 15 patients (38.5%) with moderate and mild iron overload, respectively. Nine patients (23.1%) had myocardial iron overload, which included 3 patients (7.7%) who had severe cardiac haemosiderosis. Notably, 37 patients (94.9%) had liver iron overload, which included 15 patients (38.5%) who had severe liver haemosiderosis. There was a moderate positive correlation between the relaxation time of the pancreas and heart haemosiderosis (r = 0.504, P < 0.001). No significant correlation was found between the relaxation time of the pancreas with the liver and the heart with the liver.ConclusionPancreatic haemosiderosis precedes cardiac haemosiderosis, which establishes a basis for initiating earlier iron chelation therapy to patients with thalassemia major.

Benchmarks offer a scientific way to compare algorithms using objective performance metrics. Good benchmarks have two features: (a) they should be widely useful for many research groups; (b) and they should produce reproducible findings. In robotic manipulation research, there is a trade-off between reproducibility and broad accessibility. If the benchmark is kept restrictive (fixed hardware, objects), the numbers are reproducible but the setup becomes less general. On the other hand, a benchmark could be a loose set of protocols (e.g. object set [9]) but the underlying variation in setups make the results non-reproducible. In this paper, we re-imagine benchmarking for robotic manipulation as state-of-the-art algorithmic implementations, alongside the usual set of tasks and experimental protocols. The added baseline implementations will provide a way to easily recreate SOTA numbers in a new local robotic setup, thus providing credible relative rankings between existing approaches and new work. However, these "local rankings" could vary between different setups. To resolve this issue, we build a mechanism for pooling experimental data between labs, and thus we establish a single global ranking for existing (and proposed) SOTA algorithms. Our benchmark, called Ranking-Based Robotics Benchmark (RB2), is evaluated on tasks that are inspired from clinically validated Southampton Hand Assessment Procedures [27]. Our benchmark was run across two different labs and reveals several surprising findings. For example, extremely simple baselines like open-loop behavior cloning, outperform more complicated models (e.g. closed loop, RNN, Offline-RL, etc.) that are preferred by the field. We hope our fellow researchers will use RB2 to improve their research's quality and rigor.

No abstract