Artificial Intelligence for Caries Detection: Value of Data and Information.

If increasing practitioners' diagnostic accuracy, medical artificial intelligence (AI) may lead to better treatment decisions at lower costs, while uncertainty remains around the resulting cost-effectiveness. In the present study, we assessed how enlarging the data set used for training an AI for caries detection on bitewings affects cost-effectiveness and also determined the value of information by reducing the uncertainty around other input parameters (namely, the costs of AI and the population's caries risk profile). We employed a convolutional neural network and trained it on 10%, 25%, 50%, or 100% of a labeled data set containing 29,011 teeth without and 19,760 teeth with caries lesions stemming from bitewing radiographs. We employed an established health economic modeling and analytical framework to quantify cost-effectiveness and value of information. We adopted a mixed public-private payer perspective in German health care; the health outcome was tooth retention years. A Markov model, allowing to follow posterior teeth over the lifetime of an initially 12-y-old individual, and Monte Carlo microsimulations were employed. With an increasing amount of data used to train the AI sensitivity and specificity increased nonlinearly, increasing the data set from 10% to 25% had the largest impact on accuracy and, consequently, cost-effectiveness. In the base-case scenario, AI was more effective (tooth retention for a mean [2.5%-97.5%] 62.8 [59.2-65.5] y) and less costly (378 [284-499] euros) than dentists without AI (60.4 [55.8-64.4] y; 419 [270-593] euros), with considerable uncertainty. The economic value of reducing the uncertainty around AI's accuracy or costs was limited, while information on the population's risk profile was more relevant. When developing dental AI, informed choices about the data set size may be recommended, and research toward individualized application of AI for caries detection seems warranted to optimize cost-effectiveness.

Efficiency of trapping processes in regular and disordered networks.

We use a Markov method to study the efficiency of trapping processes involving both a random walker and a deep trap in regular and disordered networks. The efficiency is gauged by the mean absorption time (average of the mean number of steps performed by the random walker before being absorbed by the trap). We compute this quantity in terms of different control parameters, namely, the length of the walker jumps, the mobility of the trap, and the degree of spatial disorder of the network. For a proper choice of the system size, we find in all cases a nonmonotonic behavior of the efficiency in terms of the corresponding control parameter. We thus arrive at the conclusion that, despite the decrease of the effective system size underlying the increase of the control parameter, the efficiency is reduced as a result of an increase of the escape probability of the walker once it finds itself in the interaction zone of the trap. This somewhat anti-intuitive effect is very robust in the sense that it is observed regardless of the specific choice of the control parameter. For the case of a ring lattice, results for the mean absorption time in systems of arbitrary size are given in terms of a two-parameter scaling function. For the case of a mobile trap, we deal with both trapping via a single channel (walker-trap overlap) and via two channels (walker-trap overlap and walker-trap crossing), thereby generalizing previous work. As for the disordered case, our analysis concerns small world networks, for which we see several crossovers of the absorption time as a function of the control parameter and the system size. The methodology used may be well suited to exploring characteristic time scales of encounter-controlled phenomena in networks with a few interacting elements and the effect of geometric constraints in nanoscale systems with a very small number of particles.

Toxicology of fires: an emerging clinical concern.

As a wider variety of synthetic materials is used in buildings, the potential for poisoning from inhalation of products of combustion is increasing greatly. Research during the past few years has shown that the burning of plastics (insulation, furniture, carpeting, electric wiring covering, decorative items) results in the formation of large amounts of highly toxic chemicals. Clinicians treating victims of fires should be aware of the toxicological ramifications of combustion, including delayed pathophysiological sequellae. A case report and a review of some current hypotheses of fire-induced toxicity illustrate the current state of knowledge, as well as some uncertainties and controversies in fire toxicology.