Despite the fact that during the period 2000–2010 road fatalities in Europe (EU27) have been reduced by 42.8% , in 2010 about 31.000 people were killed in road accidents, and about 300.000 were seriously injured. During the same period of time, Italy reduced the total number of victims by the 42.4%, but the number of injured people (light and serious) is still very high (about 300.000 in 2010) .
Vulnerable Road Users (VRU) (pedestrians, cyclists and PTW rider and pillion passenger) today are still at a very high risk of sustaining serious injuries, or being in a fatal accident, especially in metropolitan areas. Medical information on people admitted in a Tuscan Region Intensive Care Unit, and not dead on-scene of accident, is stored in the TTR [3–5] database. The 2009 and 2010 data of the TTR shows that 65% of severe injuries in the region are caused by road accidents. Twenty-nine percent of severe injuries occurred in non-urban areas and the majority (33%) in urban areas. In the metropolitan area of Florence, 62% of the severe injuries involved PTW rider and pillion passengers, 20% car occupants, 10% cyclists, and 7% pedestrians. The most frequent serious accident configurations are vehicle to vehicle accidents (73%), and run over pedestrians (18%).
Other important features are the permanent consequences sustained by people subjected to serious injury. A six-month follow-up after the traumatic event highlights that 7% of the people die, 2% remain in a vegetative state, 18% and 32% suffered, respectively, a serious and moderate disability, while 41% show a good recovery.
The analysis of the state-of-the-art shows that in-depth knowledge of real road accident data is very important for the comprehension of accident causation, mechanism of injury, and injury patterns . Today the effects of accidents on car occupants and vulnerable road users are much better known than in the past, thanks to crash tests and computer simulation techniques.
The aim of crash tests is to provide qualitative or quantitative data, the first regarding the body parts that have impacted with some area of the passenger compartment or external vehicle and the kinematic followed by the occupants, while the second regards the acceleration and force parameters on each body region of the test dummies.
All this information is useful to understand the body part injured and the computation of the injury criterions, i.e. the Head Injury Criterion (HIC) and the Neck Injury Criterion (NIC) for the head and neck regions [7, 8].
From these it is possible to calculate the probability of having a lesion corresponding to a given AIS score respective to these body regions.
In the crash tests, Anthropometric Dummies (AD) or Post Mortem Human Subject (PMHS) are used for the evaluation of the injuries, while in the computer simulation techniques Multi-Body Human Models (MBHM) or Finite-Element Human Models (FEHM) are used respectively. However, the AD and MBHM are not completely satisfactory give that the capability of the dummies to reproduce human behaviour, and particularly the injury description, is limited . But also for the PMHS, the correlation with real injuries does not always correspond to the real outcome, due to the condition of the PMHS, e.g. inactive muscles, decomposition, positioning and support, age, height, weight.
In-depth accident studies allow the monitoring of the injuries sustained by the people involved in serious road accidents, in term of type, localization, frequency and severity compared to vehicle and crash configurations, objects impacted, and so on.
This type of research gives the possibility to relate real accident situations, as well as crash tests. Structures causing injuries can be recognized at an early stage. Feedback regarding the road traffic engineering can also be obtained.
The data is also used for recognizing and assessing potential areas of future safety developments, evaluating vehicle safety performance in real world accident situations, and supporting and validating computer simulations. For example, statistical data on important factors, e.g. impact speed, angle, and mass, can be used as the basis for defining standards for impact tests.
Some of the main real world in-depth accidents studies across Europe include the “German In Depth investigation Accident Study” (GIDAS)  in Germany, the Co-operative Crash Injury Study (CCIS)  and “On The Spot” (OTS)  in the United Kingdom, the “In-Depth Car Accident Analysis” (EDA) of INRETS in France  and the SafetyNet project operating until 2008 in six European countries . For the in-depth study of road accidents focused on the PTW, the “Motorcycle Accident In depth Study” (MAIDS)  project is the reference for this type of vehicles.
In the United States the “National Accident Sampling System “(NASS)  and the “Crash Injury Research and Engineering Network“ (CIREN)  are the main in-depth accident research systems, and in Japan there is a collaborative study by “Japan Automobile Research Institute“ (JARI), Nippon Medical School Chiba Hokuso Hospital, and the “Institute for Traffic Accident Research and Data Analysis” (ITARDA) .
All this information can be useful for a wide range of fields of research such as ‘vehicle design for active and passive safety,’ ‘biomechanics,’ ‘driver behaviour,’ ‘trauma medicine,’ ‘road design,’ and so on. The data is also used for recognizing and assessing potential areas of future safety developments, evaluating vehicle safety performance in real world accident situations, and supporting and validating computer simulations. For example, statistical data on important factors, e.g. impact speed, angle, and mass, can be used as the basis for defining standards for impact tests, but also to develop new devices or shapes to mitigate the injuries, to improve current triage operations, to develop and validate new tools for the prediction of the severity of the injuries [19, 20] and to evaluate the change produced by the countermeasures adopted.
In Italy, the collection and study of in-depth real world accident data has been very time limited in the past, and completely absent in the Tuscany region. The projects conducted in Italy are the MAIDS project, led in the Pavia province between 1999 and 2001 and focusing on PTW vehicles, and the SafetyNet project conducted in the Marche region between 2004 and 2008, where all types of road accident data were collected.
Due to the importance of the data coming from this type of study and the current absence in Italy of similar research, a medical-engineering network has been created.
In the “methods” section, our modus operandi is explained, and a case study is introduced step-by-step. In the "results" section, the main results on the analysis of the road accidents currently studied are shown. In the "discussion" section, some preliminary consideration deductible from the previous results are highlighted.