The HADES project was initiated in order to develop methods for evaluating decompression tables and in particular to study the effect of diving activity and environmental factors upon decompression outcome. The database contains detailed information about over 2000 saturation dives performed on two diving vessels over a time period of six years by a stable population of divers. The experimental model system enables the detailed study of the effect of decompression procedures with changing temperatures, muscular activity and gas composition. In this model, vascular gas bubbles are used as an indication of decompression stress. Initial evaluation of the diving data indicate that a subgroup of divers performing a larger proportion of the dives, in particular dives with stopped decompressions, have a much larger incidence of decompression sickness. The experimental system is able to maintain precise control over environmental factors, initial results indicate excellent correlation with results from human dives.
No diving activity is possible without decompression. This self-evident fact is pointed out to underline the importance of this part of the diving procedure, a part that cannot be eliminated by any equipment. Over the years, considerable time and effort has been invested in trying to develop procedures that are safe and effective. The decompression schedules used in diving today have a high level of safety and will to a large degree prevent serious injury. However, recent data indicate that changes can occur both in the lungs [I] and in the brain  can be detected in individuals who never have experienced clinical decompression sickness.
The HADES project was initiated following a series of decompression incidents in one of Seaways diving ships. Without any apparent reason or change in diving procedure, a profile that had been used successfully for many years suddenly started to give an unacceptable high incidence of DCS as is seen from TABLE 1.
The tables used were linear with a decompression speed of 30 msw/24 hours to 15 msw, pp02 was .5 bar. From 15 msw to surface the speed was 13.5 msw/24 hours. Night stops were used.
The high incidence in 1988 led us to reduce the saturation decompression speed from 30 to 27 msw/24 hours, this was followed by an increase in the number of DCS incidence in 1989. A reduction in the allowable excursion distance dramatically reduced the amount of decompression problems.
TABLE 1. INCIDENCE OF DECOMPRESSION SICKNESS HeOx SATURATION 30–70 MSW(Available in full paper)
These episodes led us to perform an evaluation of our diving procedures. As no apparent reason could be found for the sudden increase in the incidence of DCS, we had to rethink some of the basic assumptions we had used when establishing our tables. We were forced to admit that some of the empirical adjustments to the tables that we and others had performed over the years had a very weak base. In fact, although decompression procedures were developed to reduce or eliminate gas bubbles, they are evaluated upon the basis of producing few acute clinical symptoms.