Machine learning in electronic-quantum-matter imaging experiments.

For centuries, the scientific discovery process has been based on systematic human observation and analysis of natural phenomena1. Today, however, automated instrumentation and large-scale data acquisition are generating datasets of such large volume and complexity as to defy conventional scientific methodology. Radically different scientific approaches are needed, and machine learning (ML) shows great promise for research fields such as materials science2-5. Given the success of ML in the analysis of synthetic data representing electronic quantum matter (EQM)6-16, the next challenge is to apply this approach to experimental data-for example, to the arrays of complex electronic-structure images17 obtained from atomic-scale visualization of EQM. Here we report the development and training of a suite of artificial neural networks (ANNs) designed to recognize different types of order hidden in such EQM image arrays. These ANNs are used to analyse an archive of experimentally derived EQM image arrays from carrier-doped copper oxide Mott insulators. In these noisy and complex data, the ANNs discover the existence of a lattice-commensurate, four-unit-cell periodic, translational-symmetry-breaking EQM state. Further, the ANNs determine that this state is unidirectional, revealing a coincident nematic EQM state. Strong-coupling theories of electronic liquid crystals18,19 are consistent with these observations.

Retrospective prevalence of snakebites from Hospital Kuala Lumpur (HKL) (1999-2003).

A hospital based retrospective study of the prevalence of snakebite cases at Hospital Kuala Lumpur was carried out over a five-year period from 1999 to 2003. A total of 126 snakebite cases were recorded. The highest admission for snakebites was recorded in 2001 (29 cases). The majority of cases were admitted for three days or less (79%). Most of the snakebite cases were reported in the 11-30 years age group (52%). The male:female ratio was 3:1. The majority of cases were Malaysians (80%, 101 cases). Of the non-Malaysians, Indonesians constituted the most (56%, 14 cases). Bites occurred most commonly on the lower limbs (49%), followed by upper limbs (45%) and on other parts of the body (6%). No fatal cases were detected and complications were scarce. In 60% (70 cases) the snake could not be identified. Of the four species of snakes that were identified, cobra (both suspected and confirmed) constituted the largest group (25%), followed by viper (10%), python (4%) and sea snake (1%). The most common clinical presentations were pain and swelling, 92% (116 cases). All patients were put on snakebite charts and their vital signs were monitored. Of the snakebite cases, 48% (61 cases) were treated with cloxacillin and 25% (32 cases) were given polyvalent snake antivenom.

Influence of co-morbidity in the interpretation of tuberculin skin reactivity in multi-ethnic adult patients with tuberculosis.

In the Malaysian setting of multi-ethnicity and high BCG coverage, interpretation of Tuberculin Skin Testing (TST) may be difficult. Between January 2001 and December 2003, a retrospective study on all adult patients with documented TST results treated for tuberculosis (TB) in chest clinics of two government hospitals was conducted to determine the reliability of TST and factors affecting its interpretation. One hundred and three patients [mean age (SD): 43 (17); male: 67%] were eligible for data collection: 72% and 57% of patients had positive TST results based on cut-off points of 10mm and 15mm respectively. The only significant univariate association with TST results was the severity of co-morbidity. A patient with co-morbidity score of 3 defined as those with any cancer, end-stage renal or liver disease, or HIV disease, was more likely to have a negative TST results [10mm cut-off point: Odd Ratio (95% CI) 6.6 (1.82 to 24.35), p = 0.003; 15mm cut-off point: 4.8 (1.21 to 18.95), p = 0.012]. A TST reading of 10mm had a higher sensitivity than 15mm as the cut-off point in diagnosing TB infection. Considering all possible confounding factors like ethnicity, prior BCG vaccination and TB burden in the population, severity of co-morbidity remains strongly predictive of a negative TST. Caution should be exercised in interpreting TST in these patients.