Partial Multiple Imputation With Variational Autoencoders: Tackling Not at Randomness in Healthcare Data.

Missing data can pose severe consequences in critical contexts, such as clinical research based on routinely collected healthcare data. This issue is usually handled with imputation strategies, but these tend to produce poor and biased results under the Missing Not At Random (MNAR) mechanism. A recent trend that has been showing promising results for MNAR is the use of generative models, particularly Variational Autoencoders.

Interactions and Supramolecular Organization of Sulfonated Indigo and Thioindigo Dyes in Layered Hydroxide Hosts.

Supramolecularly organized host-guest systems have been synthesized by intercalating water-soluble forms of indigo (indigo carmine, IC) and thioindigo (thioindigo-5,5'-disulfonate, TIS) in zinc-aluminum-layered double hydroxides (LDHs) and zinc-layered hydroxide salts (LHSs) by coprecipitation routes. The colors of the isolated powders were dark blue for hybrids containing only IC, purplish blue or dark lilac for cointercalated samples containing both dyes, and ruby/wine for hybrids containing only TIS. The as-synthesized and thermally treated materials were characterized by Fourier transform infrared, Fourier transform Raman, and nuclear magnetic resonance spectroscopies, powder X-ray diffraction, scanning electron microscopy, and elemental and thermogravimetric analyses.