Machine Learning in Healthcare

Recent advancements in Artificial Intelligence (AI) and Machine Learning (ML) technology have brought on substantial strides in predicting and identifying health emergencies, disease populations, and disease state and immune response, amongst a few. Although, skepticism remains regarding the practical application and interpretation of results from ML-based approaches in healthcare settings, the inclusion of these approaches is increasing at a rapid pace. Here we provide a brief overview of machine learning-based approaches and learning algorithms including supervised, unsupervised, and reinforcement learning along with examples. Second, we discuss the application of ML in several healthcare fields, including radiology, genetics, electronic health records, and neuroimaging. We also briefly discuss the risks and challenges of ML application to healthcare such as system privacy and ethical concerns and provide suggestions for future applications.     

Some observations on the effect of Daflon (micronized purified flavonoid fraction of Rutaceae aurantiae) in bancroftian filarial lymphoedema

BACKGROUND: Morbidity management is a core component of the global programme for the elimination of lymphatic filariasis. In a double-blind clinical trial, the tolerability and efficacy of Daflon (500 mg) + DEC (25 mg) or DEC (25 mg) alone, twice daily for 90 days, was studied in 26 patients with bancroftian filarial lymphoedema. RESULTS: None of the patients in either drug group reported any adverse reaction throughout the treatment period (90 days). Haematological and biochemical parameters were within normal limits and there was no significant difference between the pre-treatment (day 0) and post-treatment (day 90) values. The group receiving Daflon showed significant reduction in oedema volume from day 90 (140.6 PlusMinus; 18.8 ml) to day 360 (71.8 PlusMinus; 20.7 ml) compared to the pre-treatment (day 0, 198.4 PlusMinus; 16.5 ml) value. This accounted for a 63.8% reduction in oedema volume by day 360 (considering the pre-treatment (day 0) as 100%). In the DEC group, the changes in oedema volume (between day 1 and day 360) were not significant when compared to the pre-treatment (day 0) value. The percentage reduction at day 360 was only 9%, which was not significant (P > 0.05). CONCLUSION: This study has shown that Daflon (500 mg, twice a day for 90 days) is both safe and efficacious in reducing oedema volume in bancroftian filarial lymphoedema. Further clinical trials are essential for strengthening the evidence base on the role of this drug in the morbidity management of lymphatic filariasis.     

The C. elegans F-box/WD-repeat protein LIN-23 functions to limit cell division during development

In multicellular eukaryotes, a complex program of developmental signals regulates cell growth and division by controlling the synthesis, activation and degradation of G(1) cell cycle regulators. Here we describe the lin-23 gene of Caenorhabditis elegans, which is required to restrain cell proliferation in response to developmental cues. In lin-23 null mutants, all postembryonic blast cells undergo extra divisions, creating supernumerary cells that can differentiate and function normally. In contrast to the inability to regulate the extent of blast cell division in lin-23 mutants, the timing of initial cell cycle entry of blast cells is not affected. lin-23 encodes an F-box/WD-repeat protein that is orthologous to the Saccharomyces cerevisiae gene MET30, the Drosophila melanogaster gene slmb and the human gene betaTRCP, all of which function as components of SCF ubiquitin-ligase complexes. Loss of function of the Drosophila slmb gene causes the growth of ectopic appendages in a non-cell autonomous manner. In contrast, lin-23 functions cell autonomously to negatively regulate cell cycle progression, thereby allowing cell cycle exit in response to developmental signals.     

A novel solid dosage form of rifampicin and isoniazid with improved functionality

The aim of the present investigation was to develop a novel dosage form of rifampicin and isoniazid to minimize degradation of rifampicin in acidic medium and to modulate the release of rifampicin in the stomach and isoniazid in the intestine. Gastroretentive tablets of rifampicin (150 mg) were prepared by the wet granulation method using hydroxypropyl methylcellulose, calcium carbonate, and polyethylene glycol 4000. The granules and tablets of rifampicin were characterized. Hard gelatin capsules (size 4) containing a compacted mass of isoniazid (150 mg) and dicalcium phosphate (75 mg) were enteric coated. Two tablets of rifampicin and 1 capsule (size 4) of isoniazid were put into a hard gelatin capsule (size 00). The in vitro drug release and in vitro drug degradation studies were performed. Rifampicin was released over 4 hours by zero-order kinetics from the novel dosage form. More than 90% of isoniazid was released in alkaline medium in 30 minutes. The results of dissolution studies with the US Pharmacopeia XXIII method revealed that a substantial amount of rifampicin was degraded from the immediate release capsule containing rifampicin and isoniazid powder owing to drug accumulation in the dissolution vessel and also to the presence of isoniazid. The degradation of rifampicin to 3-formyl rifampicin SV (3FRSV) was arrested (3.6%–4.8% degradation of rifampicin at 4 hours) because of the minimization of physical contact between the 2 drugs and controlled release of rifampicin in acidic medium in the modified Rossett-Rice apparatus. This study concludes that the problem of rifampicin degradation can be alleviated to a certain extent by this novel dosage form. Published: August 24, 2007     

Nearest-neighbor classifier as a tool for classification of protein families

Knowledge about protein function is essential in understanding the biological processes. A specific class or family of protein shares common structural and chemical properties amongst its member sequences. The set of properties that display its unique characteristics for clearly classifying a protein sequence into its corresponding protein family needs to be studied. Our study of these important properties conducted on four major classes of proteins namely Globins, Homeoboxes, Heat Shock proteins (HSP) and Kinase have shown that frequency of twenty naturally occurring amino acids, hydrophobic content of protein, molecular weight of protein, isoelectric point of protein, secondary structure composition of amino acid residues as helices, coils and sheets and the composition of helices, coils and sheets in the secondary structure topology plays a significant role in correctly classifying the protein into its corresponding class or family as indicated by the overall efficiency of Nearest Neighbor Classifier as 84.92%.     

Sampling properties of DNA sequence data in phylogenetic analysis

We inferred phylogenetic trees from individual genes and random samples of nucleotides from the mitochondrial genomes of 10 vertebrates and compared the results to those obtained by analyzing the whole genomes. Individual genes are poor samples in that they infrequently lead to the whole-genome tree. A large number of nucleotide sites is needed to exactly determine the whole-genome tree. A relatively small number of sites, however, often results in a tree close to the whole-genome tree. We found that blocks of contiguous sites were less likely to lead to the whole-genome tree than samples composed of sites drawn individually from throughout the genome. Samples of contiguous sites are not representative of the entire genome, a condition that violates a basic assumption of the bootstrap method as it is applied in phylogenetic studies.      

Fabrication of Modified Transport Fluconazole Transdermal Spray Containing Ethyl Cellulose and Eudragit® RS100 as Film Formers

The present investigation was undertaken to fabricate modified transport fluconazole transdermal spray using ethyl cellulose and Eudragit® RS100 as film-forming polymers. Eudragit® RS100 (X ₁) and ethyl cellulose (X ₂) were selected as independent variables in 3² full factorial design, whereas drug transport in first hour (Y ₁) and the time required for 50% drug transport (Y ₂) were selected as dependent variables. Eutectic blend of camphor and menthol was used as permeation enhancer cum solvent for film-forming polymers. The pH, viscosity, volume of solution delivered upon each actuation, spray angle, ex–in vivo physical evaluation and in vitro drug transport of the formulated products were evaluated. The optimized batch B16 containing 5.25% w/w ethyl cellulose and 10.6% w/w Eudragit® RS100 was formulated by overlapping the contour plots of Y ₁ and Y ₂. The pH, viscosity, volume of solution sprayed upon each actuation and spray angle of the batch B16 was 6.3, 52.9 cPs, 0.24 ml and 82.6° respectively. The film of optimized batch was flexible and dermal-adhesive. The responses Y ₁ and Y ₂ of batch B16 were 7.91 μg/ml and 347 min respectively. The kinetics of drug transport was best explained by the Korsmeyer and Peppas model. The eutectic mixture consisting of equal parts of camphor and menthol showed improved drug permeation through shed snake skin. Short-term stability study demonstrated insignificant changes in performance characteristics.