Computer vision in aseptic pharmaceutical manufacturing: a deep learning-based approach to vial inspection and pack formation during a lyophilizer loading cycle

Vision systems are powerful tools playing an increasingly important role in modern industry, to detect errors and maintain product standards. With the enlarged availability of affordable industrial cameras, computer vision algorithms have been increasingly applied in industrial manufacturing processes monitoring. Until a few years ago, industrial computer vision applications relied only on ad-hoc algorithms designed for the specific object and acquisition setup being monitored, with a strong focus on co-designing the acquisition and processing pipeline. Deep learning has overcome these limits providing greater flexibility and faster re-configuration. In this work, the process to be inspected consists in vials’ pack formation entering a freeze-dryer, which is a common scenario in pharmaceutical active ingredient packaging lines. To ensure that the machine produces proper packs, a vision system is installed at the entrance of the freeze-dryer to detect eventual anomalies with execution times compatible with the production specifications. Other constraints come from sterility and safety standards required in pharmaceutical manufacturing. This work presents an overview about the production line, with particular focus on the vision system designed, and about all trials conducted to obtain the final performance. Transfer learning, alleviating the requirement for a large number of training data, combined with data augmentation methods, consisting in the generation of synthetic images, were used to effectively increase the performances while reducing the cost of data acquisition and annotation. The proposed vision algorithm is composed by two main subtasks, designed respectively to vials counting and discrepancy detection. The first one was trained on more than 23k vials (about 300 images) and tested on 5k more (about 75 images), whereas 60 training images and 52 testing images were used for the second one.

Development of in vitro methods to test acetabular prosthetic reconstructions (messa a punto di metodi in vitro per la caratterizzazione biomeccanica di ricostruzioni acetabolari)

The revision hip arthroplasty is a surgical procedure, consisting in the reconstruction of the hip joint through the replacement of the damaged hip prosthesis. Several factors may give raise to the failure of the artificial device: aseptic loosening, infection and dislocation represent the principal causes of failure worldwide. The main effect is the raise of bone defects in the region closest to the prosthesis that weaken the bone structure for the biological fixation of the new artificial hip. For this reason bone reconstruction is necessary before the surgical revision operation. This work is born by the necessity to test the effects of bone reconstruction due to particular bone defects in the acetabulum, after the hip prosthesis revision. In order to perform biomechanical in vitro tests on hip prosthesis implanted in human pelvis or hemipelvis a practical definition of a reference frame for these kind of bone specimens is required. The aim of the current study is to create a repeatable protocol to align hemipelvic samples in the testing machine, that relies on a reference system based on anatomical landmarks on the human pelvis. In chapter 1 a general overview of the human pelvic bone is presented: anatomy, bone structure, loads and the principal devices for hip joint replacement. The purpose of chapters 2 is to identify the most common causes of the revision hip arthroplasty, analysing data from the most reliable orthopaedic registries in the world. Chapter 3 presents an overview of the most used classifications for acetabular bone defects and fractures and the most common techniques for acetabular and bone reconstruction. After a critical review of the scientific literature about reference frames for human pelvis, in chapter 4, the definition of a new reference frame is proposed. Based on this reference frame, the alignment protocol for the human hemipelvis is presented as well as the statistical analysis that confirm the good repeatability of the method.

Polyhydroxyalkanoates production from salted food waste using mixed microbial cultures.

Plastic is an essential asset for the modern lifestyle, given its superiority as a material from the points of view of cost, processability and functional properties. However, plastic-related environmental pollution has become nowadays a very significant problem that can no longer be overlooked. For this reason, in recent decades, the research for new materials that could replace fossil fuel-based plastics has been focused on biopolymers with similar physicochemical properties to fossil fuel-based plastics, such as Polyhydroxyalkanoates (PHA). PHAs are a family of biodegradable polyesters synthesized by many microorganisms as carbon and energy reserves. PHA appears as a good candidate to substitute conventional petroleum-based plastics since it has similar properties, but with the advantage of being biobased and biodegradable, and has a wide range of applications (e.g., packaging). However, the PHA production cost is almost four times higher (€5/kg) than conventional plastic manufacturing. The PHA production by mixed microbial cultures (MMC) allows to reduce production costs as it does not require aseptic conditions and it enables the use of inexpensive by-products or waste streams as these cultures are more amenable to deal with complex feedstocks. Saline wastewaters (WWs), generated by several industries such as seafood, leather and dairy, are often rich in organic compounds and, due to a strong salt inhibition, the biological treatments are inefficient, and their disposal is expensive. These saline WWs are a potential feedstock for PHA production, as they are an inexpensive raw material. Moreover, saline WWs could allow the utilization of seawater in the process as dilution and cleaning agent, further decreasing the operational costs and the environmental burden of the process. The main goal of the current project is to assess and optimize the PHA production from a mixture of food waste and brine wastewater from the fishery industry by MMC.