Requirements analysis for an AI-based clinical decision support system for general practitioners: a user-centered design process

BACKGROUND

As the first point of contact for patients with health issues, general practitioners (GPs) are frequently confronted with patients presenting with non-specific symptoms of unclear origin. This can result in delayed, prolonged or false diagnoses. To accelerate and improve the diagnosis of diseases, clinical decision support systems would appear to be an appropriate tool. The objective of the project 'Smart physician portal for patients with unclear disease' (SATURN) is to employ a user-centered design process based on the requirements analysis presented in this paper to develop an artificial Intelligence (AI)-based diagnosis support system that specifically addresses the needs of German GPs.

METHODS

Requirements analysis for a GP-specific diagnosis support system was conducted in an iterative process with five GPs. First, interviews were conducted to analyze current workflows and the use of digital applications in cases of diagnostic uncertainty (as-is situation). Second, we focused on collecting and prioritizing tasks to be performed by an ideal smart physician portal (to-be situation) in a workshop. We then developed a task model with corresponding user requirements.

RESULTS

Numerous GP-specific user requirements were identified concerning the tasks and subtasks: performing data entry (open system, enter patient data), reviewing results (receiving and evaluating results), discussing results (with patients and colleagues), scheduling further diagnostic procedures, referring to specialists (select, contact, make appointments), and case closure. Suggested features particularly concerned the process of screening and assessing results: e.g., the system should focus more on atypical patterns of common diseases than on rare diseases only, display probabilities of differential diagnoses, ensure sources and results are transparent, and mark diagnoses that have already been ruled out. Moreover, establishing a means of using the platform to communicate with colleagues and transferring patient data directly from electronic patient records to the system was strongly recommended.

CONCLUSIONS

Essential user requirements to be considered in the development and design of a diagnosis system for primary care could be derived from the analysis. They form the basis for mockup-development and system engineering.

Low antibody prevalence against Bacillus cereus biovar anthracis in Taï National Park, Côte d'Ivoire, indicates high rate of lethal infections in wildlife

Bacillus cereus biovar anthracis (Bcbva) is a member of the B. cereus group which carries both B. anthracis virulence plasmids, causes anthrax-like disease in various wildlife species and was described in several sub-Saharan African rainforests. Long-term monitoring of carcasses in Taï National Park, Côte d'Ivoire, revealed continuous wildlife mortality due to Bcbva in a broad range of mammalian species. While non-lethal anthrax infections in wildlife have been described for B. anthracis, nothing is known about the odds of survival following an anthrax infection caused by Bcbva. To address this gap, we present the results of a serological study of anthrax in five wildlife species known to succumb to Bcbva in this ecosystem. Specific antibodies were only detected in two out of 15 wild red colobus monkeys (Procolobus badius) and one out of 10 black-and-white colobus monkeys (Colobus polykomos), but in none of 16 sooty mangabeys (Cercocebus atys), 9 chimpanzees (Pan troglodytes verus) and 9 Maxwell's duikers (Cephalophus maxwellii). The combination of high mortality and low antibody detection rates indicates high virulence of this disease across these different mammalian species.

Comparative analysis of the immunologic response induced by the Sterne 34F2 live spore Bacillus anthracis vaccine in a ruminant model

The Sterne 34F2 live spore vaccine (SLSV) developed in 1937 is the most widely used veterinary vaccine against anthrax. However, literature on the immunogenicity of this vaccine in a target ruminant host is scarce. In this study, we evaluated the humoral response to the Bacillus anthracis protective antigen (rPA), a recombinant bacillus collagen-like protein of anthracis (rBclA), formaldehyde inactivated spores (FIS) prepared from strain 34F2 and a vegetative antigen formulation prepared from a capsule and toxin deficient strain (CDC 1014) in Boer goats. The toxin neutralizing ability of induced antibodies was evaluated using an in vitro toxin neutralization assay. The protection afforded by the vaccine was also assessed in vaccinates. Anti-rPA, anti-FIS and lethal toxin neutralizing titres were superior after booster vaccinations, compared to single vaccinations. Qualitative analysis of humoral responses to rPA, rBclA and FIS antigens revealed a preponderance of anti-FIS IgG titres following either single or double vaccinations with the SLSV. Antibodies against FIS and rPA both increased by 350 and 300-fold following revaccinations respectively. There was no response to rBclA following vaccinations with the SLSV. Toxin neutralizing titres increased by 80-fold after single vaccination and 700-fold following a double vaccination. Lethal challenge studies in naïve goats indicated a minimum infective dose of 36 B. anthracis spores. Single and double vaccination with the SLSV protected 4/5 and 3/3 of goats challenged with>800 spores respectively. An early booster vaccination following the first immunization is suggested in order to achieve a robust immunity. Results from this study indicate that this crucial second vaccination can be administered as early as 3 months after the initial vaccination.

BclA and toxin antigens augment each other to protect NMRI mice from lethal Bacillus anthracis challenge

While proving highly effective in controlling Anthrax in farm animals all over the world currently attenuated live anthrax vaccines employed in a veterinary context suffer from drawbacks such as residual virulence, short term protection, variation in quality and, most importantly, lack of efficacy if administered simultaneously with antibiotics. These limitations have stimulated the development of non-living component vaccines which induce a broad spectrum immune response capable of targeting both toxaemia (as in the case of PA based vaccines) and bacteraemia. To contribute to this several new approaches were tested in outbred NMRI mice for antibody titres and protectiveness. Plasmids encoding a recombinant toxin derived fusion peptide and a spore surface derived peptide were tested as DNA-vaccines in comparison to their protein counterparts utilising two adjuvant approaches and two DNA-vector backbones. The combination of two plasmids encoding LFD1PAD4-mIPS1 and TPA-BclAD1D3-LAMP1, when delivered by GeneGun, protected 90% of the animals against a lethal challenge with 25LD50 spores of the Ames strain of Bacillus anthracis. Single applications of either antigen component showed significantly lower protection rates, indicating the beneficial interaction between anti-spore and anti-toxin components for an acellular vaccine formulation.

Synthesis and immunochemical evaluation of a non-methylated disaccharide analogue of the anthrax tetrasaccharide

Anthrax tetrasaccharide is an oligosaccharide expressed at the outermost surface of the Bacillus anthracis spores, featuring three rhamnoses and a rare sugar called anthrose. This motif has now been identified as a plausible component of future human vaccines against anthrax. We report herein the synthesis of a 2-O-demethylated-β-D-anthropyranosyl-(1→3)-α-L-rhamnopyranose disaccharide analogue of this tetrasaccharide from a cyclic sulfate intermediate. This disaccharide conjugated to BSA induces an anti-native tetrasaccharide IgG antibody response when administered in BALB/c mice. Moreover, induced sera bound to native B. anthracis endospores. These results suggest that the disaccharide analogue, easily amenable for a synthetic scale-up, could be used in a glycoconjugate antigen formulation.