Nowadays, according to the World Health Organization (WHO), malaria is one of the most common life-threatening infectious diseases and a global public health challenge. It is caused by microscopic Plasmodium parasites that are transmitted to people through the bites of infected Anopheles mosquitoes, also called “malaria vectors” [1, 2]. In 2016, around 216 million cases were registered with an estimated 445000 deaths worldwide. The majority of these, around 90%, occurred in the sub-Saharan Africa where the population exposed to the highest risk of death is composed by children under 5 that are more susceptible to infection and illness because of their low developed immune system. Fighting malaria requires a prompt and efficient diagnosis which leads to an immediate treatment [1,6]. Even though a simple clinical diagnosis, based only on symptoms evaluation, is by far the cheapest strategy and widely used in the past, the differentiation between malaria and other febrile conditions might result complicated, leading to an indiscriminate use of anti-malarial drugs. In addition, the accuracy of the diagnosis varies with age group, season and endemicity of the interested zone. At present, the most relevant methodologies for malaria diagnosis concern therefore the direct parasite investigation by a microscopic examination (gold standard) and the indirect antigen detection implemented by rapid diagnostic tests (RDTs). The resolution (i.e. minimum detectable parasitaemia) of the microscope-based techniques ranges from 4-20parasites/μL in a clinical condition and 50-100parasites/μL for an on-field diagnosis. The sensitivity and specificity depend on the microscopist ability. Nevertheless, it requires a healthcare facility able to sustain a time spending process (≈60 minutes) with trained personnel. RDTs, instead, allows for a faster diagnosis (10-15 minutes) with a declared limit of detection of around 50-100parasites/μL and a sensitivity greater than 95%. However, this sensitivity lowers in case of non-p.falciparum plasmodium species and the diagnosis might be altered by extreme temperature and humidity conditions. Moreover, based on the nature of the test, RDTs are limited in the quantification of the parasitaemia [22,23]. This thesis work is part of the TID-MEKII project which aims at developing a compact, low cost and easy point-of-care diagnostic system, which allows a pan-plasmodic (i.e. all plasmodium species) and rapid malaria detection, as sensitive and accurate as the gold standard. The physical concept is that of exploiting the paramagnetic property of malaria infected erythrocytes and hemozoin crystals [14,17,19,21,28], which allows them to undergo a selective magnetophoretic separation driven by a magnetic field gradient. Upon separation, they concentrate at the surface of gold interdigitated measurement electrodes placed in close proximity to the magnetic concentrators whereas, healthy ones sediment under the action of gravity. Then, a change in resistivity, proportional to the amount of attracted particles, can be detected as an impedance variation by an electronic circuit. Following a lab-on-chip approach, the diagnosis is performed thanks to a disposable chip composed by an array of micrometric nickel (Ni) pillars, integrated in a silicon (Si) wafer chip with planar interdigitated gold electrodes above them. The blood droplet, which has already been diluted in a solution of PBS (Phosphate Buffer Saline) and anticoagulant solution to prevent clogging, is placed on a disposable glass petri dish. The chip is then put in close contact to the glass slide, at a distance defined by an outer ring (gasket), which identifies also the blood volume where detection takes place. The required magnetic field gradient, able to sustain the magnetophoretic separation, is provided macroscopically by a system of permanent magnets and microscopically by the Nickel pillars. The magnets are mounted on a reusable module and lowered vertically until they are in contact with the back of the chip. The resulting magnetic field gradient allows the paramagnetic particles to be attracted vertically, overcoming the drag and the gravity force. To perform the electrical readout in a reliable way, minimizing the signal to noise ratio (SNR), the system computes the difference between the signal coming from the working electrodes and the reference electrodes. The former are fabricated above the Ni concentrators and then are able to attract the particles whereas, the latter are fabricated directly onto the silicon substrate, sensing therefore the signal provided by the medium solution only. TID-MEKII provides therefore an effective alternative to the current technologies for the diagnosis of the whole plasmodium parasithosis (pan-plasmodic) as well as for the patient outcome evaluation or follow-up. The progressive increase in the number of hemozoin crystals [20,21], and the increased number of infected RBCs over the intra-erythrocytic stage, allows, in fact, to quantify the parasitaemia level. The expected time required to perform the test is around 5 minutes and this would represent a great advantage with respect to the state of the art of microscopy techniques and RDTs. Moreover, the expected specificity is around 100% with an absence of false positive outcomes and the resolution should be comparable to that of a microscopic approach; around 100 parasites/μL. Despite these ideal features, some deviations might arise in a practical use and their evaluation is part of this thesis work. Finally, in view of an on-field use in endemic and poor developing realities, the system should be as simple as possible in order not to require any specialized personnel. The capability of the device of performing the magnetophoretic separation has been tested by means of capture experiments. The majority of these experiments have been conducted with treated bovine red blood cells (t-RBCs) stained with a fluorescent marker based on the green fluorescent protein (GFP). The treatment consists of turning the oxyhemoglobin, which has a diamagnetic property, into methemoglobin (i.e. the oxidized form of hemoglobin) that, instead, is paramagnetic [62]. The geometry of attraction required an inverted microscope and the fluorescence allowed to accurately count the cells, evaluating therefore the “capture efficiency” as a ratio between the attracted t-RBCs and the total t-RBCs in each field of view. The evaluation of this capture efficiency ratio, allowed to follow the dynamic of attraction over time and in different conditions. A comparison between the bovine and the human blood has been firstly performed, showing an identical behavior, ensuring therefore a significance of the further experiments. The experimental trials aimed at determining the dynamic of attraction in different buffer solutions (i.e. different dilutions) and compositions (i.e. with or without healthy RBCs matrix with different hematocrit levels). A dilution of 10 times in PBS (Phosphate Buffer Saline) has been chosen since it allowed a reliable count of the stained RBCs. One of the goals of these experiments was the determination of the best layout for the Ni array among different geometries. The best geometry composes of a hexagonal layout of Ni cilinders, fabricated into the silicon substrate, of 20µm height, a diameter of 40µm and a relative spacing between their centres of 160µm. This geometry has highlighted an achievable level of capture efficiency, within 15 minutes, of around 75% in a suspension of RBCs in plasma and PBS for a total hematocrit value of 4% (Hct), where the volumetric fraction of infected RBCs was 0,02% (Ft). This capture efficiency has been improved by applying an active vibrating stimulus to the system, reaching a value of 85% after only 5 minutes. In the same conditions, the false positive outcomes (a-specific capture efficiency) has been evaluated to be null in the first 10 minutes and few percentage points in the following 5 minutes, probably due to the fluorescent dye locally released in solution. A final experiment has demonstrated the possibility of working also on the “whole blood” since the platelets and the white blood cells does not affect the capture outcome that much. To conclude, these tests have proven the feasibility of the magnetophoretic separation with a specificity of 100% in the first 5 minutes. If confirmed also for real human infected blood, this aspect would represent a great advantage with respect to the state of the art of microscopic technique and RDTs. Based on the previous results, a perfect chip active area, as well as the optimal geometrical layout for the connection to the reusable module, have been designed. Several detection experiments have been therefore conducted with both treated bovine erythrocytes and hemozoin crystals. The former gave a positive result with an inverted capture geometry (i.e. sedimentation) whereas, the latter were positively detected also in the direct geometry, providing a calibration curve up to a concentration of crystals equal to 1,25µg/µL. In both cases, however, an improvement in the resolution of the electronic readout system needs to be accomplished for achieving the limit of detection of the current RDTs and microscope-based diagnostic technologies. This thesis work concerns the development of the silicon chip fabrication process as well as the experiments of magnetophoretic capture conducted in order to optimize the disposable chip, in terms of concentrators and electrodes, the geometrical layout and the connection to the reusable module. Moreover, the theoretical limit of detection (LoD) of this technology is investigated and some preliminar impedance detection measurement are performed. The silicon chip has been realized at PoliFab, the micro- and nano-fabrication facility of “Politecnico di Milano” with the standard photolithographic processes, exploiting cleanroom microfabrication equipment and techniques. The experimental activities have been performed under the supervision of Prof. Riccardo Bertacco, head of the Nanomagnetism group of the Department of Physics. The bovine and human blood samples used in the experiments, along with the required preparation, were provided by “µBS lab” of “Politecnico di Milano” and some of the magnetophoretic validation experiments where conducted in the “Mechanobiology lab”. The integrated electronic board that should allow the differential measure directly onto the module has been designed, in parallel, in the framework of another MS thesis work of the "I3N group” of “Politecnico di Milano”. The results achieved pave the way for a further experimental campaign at “Luigi Sacco” hospital in Milan in view of a second and most relevant on-field experimental campaign directly in the endemic area of Youndeè (capital of Camerun) at the “Centre Medical jean Zoa” hospital.
Al giorno d'oggi, secondo l'organizzazione mondiale della sanità (OMS), la malaria è una tra le parassitosi più pericolose e rappresenta un pericolo a livello mondiale in ambito sanitario. Essa è causata dal plasmodio, un parassita che viene trasmesso agli esseri umani attraverso la puntura di una zanzara Anopheles [1,2]. Nel 2016, sono stati registrati nel mondo circa 216 milioni di casi per un totale di 445000 decessi, la maggior parte dei quali è avvenuta nelle regioni dell'Africa subsahariana che vede nei bambini sotto il quinto anno di età la fetta di popolazione maggiormente esposta al rischio di morte. La lotta alla malaria richiede una diagnosi rapida ed efficiente in modo quindi da somministrare le cure necessarie il prima possibile e solamente nei casi in cui siano realmente necessarie, limitando lo sviluppo di resistenza ai farmaci da parte del parassita [1,6]. Nonostante la diagnosi clinica, basata solamente sull'analisi della sintomatologia, sia stata la tecnica più utilizzata in passato, la discriminazione tra una febbre malarica ed altre condizioni febbrili potrebbe risultare complicata e condurrebbe, in alcuni casi, ad una inutile somministrazione di farmaci antimalarici, aumentando la resistenza agli stessi da parte del parassita. Inoltre, l'accuratezza della diagnosi varia in base all'età, alla stagione e alla endemicità della località geografica di interesse. Le tecnologie più rilevanti per la diagnosi della malaria riguardano la ricerca diretta del parassita attraverso un'analisi microscopica (gold standard) e la rilevazione indiretta degli antigeni prodotti dal plasmodio (implementata dai test diagnostici rapidi; RDT). Il limite di rilevazione (Limit of Detection - LOD), ovvero il minimo livello di parassitemia rilevabile, del test microscopico varia da un minimo di 4-20 parassiti/µL per un test svolto in ambiente clinico a 50-100 parassiti/µL per un test svolto direttamente sul campo da un microscopista medio, dal quale dipende anche il livello di sensibilità e specificità del test. Tuttavia, le tecniche microscopiche necessitano di una struttura sanitaria con personale specializzato in grado di sostenere un processo dispendioso in termini di tempo (circa 60 minuti a test). I test diagnostici rapidi, al contrario, permettono una diagnosi più rapida (10-15 minuti per test) con un limite di detezione dichiarato di circa 50-100 parassiti/µL e una sensibilità maggiore del 95%. Quest'ultima, tuttavia, è ridotta in caso di specie di plasmodio diverse dal falciparum e, più in generale, la diagnosi potrebbe essere alterata da condizioni ambientali estreme (temperatura e umidità). In aggiunta, per la natura del test stesso, i RDTs non sono in grado di fornire una quantificazione del livello di parassitemia [22,23]. Questo lavoro di tesi è parte del progetto TID-MEKII, il cui obiettivo è quello di realizzare un dispositivo diagnostico rapido, semplice e a basso costo che permetta una diagnosi sul campo di tutti i principali ceppi della malaria (pan-plasmodico) che raggiunga un livello di accuratezza pari al gold standard. Il concetto fisico alla base del progetto è quello di sfruttare le proprietà paramagnetiche dei cristalli di emozoina che si formano all'interno dei globuli rossi e che ne determinano un comportamento "paramagnetico" rispetto al mezzo (plasma) in cui sono immersi [14,17,19,20,21,28]. L'applicazione di un gradiente di campo magnetico sufficientemente elevato comporta quindi una separazione tra i globuli “sani” e i globuli “infetti”; gli infetti vengono attratti in prossimità di elettrodi interdigitati realizzati sulla sommità di concentratori magnetici, mentre i sani sedimentano sotto l'azione della gravità. Una variazione di impedenza, proporzionale al numero di particelle attratte, viene quindi rilevata. Seguendo un approccio di tipo “lab-on-chip”, la diagnosi è resa possibile grazie ad un chip di Silicio (Si) composto da un pattern di strutture cilindriche micrometriche in Nickel (Ni) integrate nel substrato stesso, sulla sommità delle quali sono realizzati gli elettrodi interdigitati d'oro (Au). La goccia di sangue, diluito in una soluzione di PBS (Phosphate Buffer Saline) e anticoagulante, è posizionata su un vetrino sul quale poi viene adagiato il chip. La distanza tra vetrino e chip è definita da una struttura contenitiva saldata al vetrino e alta 40µm, che definisce quindi il volume sensibile sul quale avviene la misura. Il gradiente di campo magnetico richiesto per sostenere la separazione magnetoforetica è creato macroscopicamente da un sistema di magneti permanenti e microscopicamente dai concentratori di Nickel. I magneti sono montati esternamente su un modulo riutilizzabile e abbassati verticalmente fino al contatto con il retro del chip. A questo punto i concentratori sono magnetizzati grazie al campo esterno e sono in grado di attrarre le particelle verticalmente, vincendo la forza di gravità e la forza di attrito viscoso. Al fine di ridurre il rapporto segnale / rumore (SNR) il sistema analizza la differenza tra il segnale rilevato da elettrodi di misura, fabbricati sopra i concentratori magnetici, e un segnale ottenuto da elettrodi di riferimento, i quali non sono provvisti di un relativo concentratore magnetico, fornendo di conseguenza un segnale dato interamente dalla soluzione. TID-MEKII si pone quindi come obiettivo quello di fornire un'alternativa alle esistenti tecnologie per la diagnosi di tutte le specie di plasmodio (pan-plasmodico) così come quello di fornire una valutazione delle condizioni del paziente durante tutto il processo di cura della patologia. Il tempo atteso per la diagnosi è pari a 5 minuti, rappresentando un vantaggio rispetto allo stato dell'arte. Inoltre, la sensibilità e specificità attese sono dell'ordine del 100%, senza quindi la presenza di falsi positivi. La risoluzione che il dispositivo si propone di raggiungere è comparabile con quella dei test diagnostici rapidi e test microscopici (≈100 parassiti/µL [22,23]). Tuttavia, l’utilizzo nella pratica potrebbe introdurre delle incertezze e conseguenti deviazioni da questi parametri ideali. Infine, in vista di un possibile uso direttamente presso le zone endemiche dei paesi in via di sviluppo, il sistema si pone come obiettivo quello di essere il più semplice possibile, in modo quindi da non richiedere necessariamente la presenza di personale qualificato. Gli esperimenti di separazione magnetoforetica sono stati condotti mediante l'utilizzo di globuli rossi bovini, trattati e colorati con una soluzione fluorescente basata sulla proteina GFP (green fluorescent protein). Il trattamento consiste nel trasformare l'emoglobina dei globuli rossi ossigenati (ossiemoglobina), che presenta proprietà diamagnetiche, in metemoglobina che, al contrario, è paramagnetica [62]. Data la geometria di cattura del dispositivo, durante questi esperimenti è stato utilizzato un microscopio invertito che potesse lavorare in fluorescenza, rendendo quindi possibile una conta delle cellule e la successiva valutazione dell'efficienza di cattura, calcolata come rapporto tra i globuli trattati attratti e i globuli trattati totali. Il calcolo di questo fattore ha reso possibile una analisi dettagliata della dinamica di attrazione nel tempo e in diverse condizioni sperimentali. Un primo fondamentale esperimento ha illustrato lo stesso andamento nella dinamica di cattura dei globuli umani e dei globuli bovini, permettendo quindi l'utilizzo degli ultimi come modello del comportamento reale. I successivi esperimenti hanno indagato la cattura dei globuli in percentuali differenti, risospesi in differenti soluzioni e livelli di ematocrito. Per garantire una conta accurata delle cellule attratte è stata scelta una diluizione del sangue (1/10) in PBS. In aggiunta, uno dei principali obiettivi di questi test è stato quello di determinare il layout migliore per la disposizione dei concentratori di Nickel. La geometria che garantisce una cattura ottimale vede i concentratori cilindrici di altezza pari a 20µm, un diametro pari a 40µm e spaziati di 160µm, disposti esagonalmente nel substrato di silicio. Questa geometria ha permesso di ottenere una efficienza di cattura pari a circa 75% in 15 minuti con un ematocrito (Hct) pari al 4%, dove la frazione volumetrica dei globuli trattati (Ft) era 0,02%. Questa cattura è stata migliorata attraverso l'applicazione di una vibrazione al sistema, raggiungendo un valore di circa 85% dopo solo 5 minuti. Nelle stesse condizioni, l'efficienza di attrazione dei globuli non trattati (falsi positivi) risulta essere nulla per i primi 10 minuti di test, raggiungendo un valore di pochi punti percentuale nei rimanenti 5 minuti. Un esperimento aggiuntivo ha illustrato inoltre la possibilità di lavorare su sangue intero, dimostrando che le piastrine e i globuli bianchi in soluzione non compromettono l'efficienza di cattura. Per concludere, questi esperimenti dimostrano la possibilità di separare magnetoforeticamente i globuli sani da quelli infetti con una specificità del 100% nei primi 5 minuti. Questo aspetto, se confermato su sangue umano infetto da malaria, costituirebbe un grosso vantaggio rispetto ai dispositivi diagnostici attualmente in commercio. Secondo i risultati precedenti, è stato progettato un chip con un layout ottimale in termini di area sensibile, geometria degli elettrodi e geometria dei pad in grado di permettere il contatto elettrico con il modulo riutilizzabile. Diversi esperimenti di lettura impedenziometrica sono stati svolti sia con i globuli rossi bovini trattati che con i cristalli di emozoina. I globuli rossi hanno fornito un esito positivo solamente con una geometria di cattura invertita (sedimentazione dei globuli) mentre invece i cristalli di emozoina sono stati rilevati anche in geometria corretta fino ad un valore di concentrazione pari a 1,25 μg/μL. In entrambi i casi, tuttavia, un miglioramento nella risoluzione della lettura elettronica deve essere conseguito per raggiungere il limite di detezione di un test diagnostico rapido e di un test microscopico. Questo lavoro di tesi, in particolare, riguarda lo sviluppo e la progettazione del chip di silicio così come la realizzazione degli esperimenti di ottimizzazione dello stesso, in termini di concentratori ed elettrodi, il layout geometrico e la connessione con il modulo esterno riutilizzabile. Inoltre, è stato svolto uno studio teorico sulla risoluzione dello strumento e una sessione sperimentale preliminare per analizzare il limite di detezione ottenuto. Il chip di silicio è stato realizzato presso “PoliFab”, la facility di micro- e nano-fabbricazione del “Politecnico di Milano”. L'attività sperimentale è stata svolta sotto la supervisione del Prof. Riccardo Bertacco, responsabile del gruppo di Nanomagnetismo del Dipartimento di Fisica. I campioni di sangue bovino e umano usati durante gli esperimenti, e i relativi processi di preparazione, sono stati forniti dal “µBS Lab” e parte degli esperimenti di cattura magnetoforetica sono stati condotti nel laboratorio “Mechanobiology lab”. La scheda elettronica che deve essere implementata direttamente sul modulo è stata realizzata in parallelo in un altro progetto di tesi del laboratorio " I3N group”. I risultati ottenuti aprono le porte ad un ulteriore sessione sperimentale per valutare il comportamento dei veri globuli umani infetti da malaria presso l'ospedale “Luigi Sacco” di Milano, in vista di una seconda campagna sperimentale nella zona endemica di Youndeè (capitale del Camerun) presso l'ospedale “Centre Medical jean Zoa”.
On-chip magnetophoretic capture and detection of red blood cells infected by malaria
GIULIANI, ENRICO
2017/2018
Abstract
Nowadays, according to the World Health Organization (WHO), malaria is one of the most common life-threatening infectious diseases and a global public health challenge. It is caused by microscopic Plasmodium parasites that are transmitted to people through the bites of infected Anopheles mosquitoes, also called “malaria vectors” [1, 2]. In 2016, around 216 million cases were registered with an estimated 445000 deaths worldwide. The majority of these, around 90%, occurred in the sub-Saharan Africa where the population exposed to the highest risk of death is composed by children under 5 that are more susceptible to infection and illness because of their low developed immune system. Fighting malaria requires a prompt and efficient diagnosis which leads to an immediate treatment [1,6]. Even though a simple clinical diagnosis, based only on symptoms evaluation, is by far the cheapest strategy and widely used in the past, the differentiation between malaria and other febrile conditions might result complicated, leading to an indiscriminate use of anti-malarial drugs. In addition, the accuracy of the diagnosis varies with age group, season and endemicity of the interested zone. At present, the most relevant methodologies for malaria diagnosis concern therefore the direct parasite investigation by a microscopic examination (gold standard) and the indirect antigen detection implemented by rapid diagnostic tests (RDTs). The resolution (i.e. minimum detectable parasitaemia) of the microscope-based techniques ranges from 4-20parasites/μL in a clinical condition and 50-100parasites/μL for an on-field diagnosis. The sensitivity and specificity depend on the microscopist ability. Nevertheless, it requires a healthcare facility able to sustain a time spending process (≈60 minutes) with trained personnel. RDTs, instead, allows for a faster diagnosis (10-15 minutes) with a declared limit of detection of around 50-100parasites/μL and a sensitivity greater than 95%. However, this sensitivity lowers in case of non-p.falciparum plasmodium species and the diagnosis might be altered by extreme temperature and humidity conditions. Moreover, based on the nature of the test, RDTs are limited in the quantification of the parasitaemia [22,23]. This thesis work is part of the TID-MEKII project which aims at developing a compact, low cost and easy point-of-care diagnostic system, which allows a pan-plasmodic (i.e. all plasmodium species) and rapid malaria detection, as sensitive and accurate as the gold standard. The physical concept is that of exploiting the paramagnetic property of malaria infected erythrocytes and hemozoin crystals [14,17,19,21,28], which allows them to undergo a selective magnetophoretic separation driven by a magnetic field gradient. Upon separation, they concentrate at the surface of gold interdigitated measurement electrodes placed in close proximity to the magnetic concentrators whereas, healthy ones sediment under the action of gravity. Then, a change in resistivity, proportional to the amount of attracted particles, can be detected as an impedance variation by an electronic circuit. Following a lab-on-chip approach, the diagnosis is performed thanks to a disposable chip composed by an array of micrometric nickel (Ni) pillars, integrated in a silicon (Si) wafer chip with planar interdigitated gold electrodes above them. The blood droplet, which has already been diluted in a solution of PBS (Phosphate Buffer Saline) and anticoagulant solution to prevent clogging, is placed on a disposable glass petri dish. The chip is then put in close contact to the glass slide, at a distance defined by an outer ring (gasket), which identifies also the blood volume where detection takes place. The required magnetic field gradient, able to sustain the magnetophoretic separation, is provided macroscopically by a system of permanent magnets and microscopically by the Nickel pillars. The magnets are mounted on a reusable module and lowered vertically until they are in contact with the back of the chip. The resulting magnetic field gradient allows the paramagnetic particles to be attracted vertically, overcoming the drag and the gravity force. To perform the electrical readout in a reliable way, minimizing the signal to noise ratio (SNR), the system computes the difference between the signal coming from the working electrodes and the reference electrodes. The former are fabricated above the Ni concentrators and then are able to attract the particles whereas, the latter are fabricated directly onto the silicon substrate, sensing therefore the signal provided by the medium solution only. TID-MEKII provides therefore an effective alternative to the current technologies for the diagnosis of the whole plasmodium parasithosis (pan-plasmodic) as well as for the patient outcome evaluation or follow-up. The progressive increase in the number of hemozoin crystals [20,21], and the increased number of infected RBCs over the intra-erythrocytic stage, allows, in fact, to quantify the parasitaemia level. The expected time required to perform the test is around 5 minutes and this would represent a great advantage with respect to the state of the art of microscopy techniques and RDTs. Moreover, the expected specificity is around 100% with an absence of false positive outcomes and the resolution should be comparable to that of a microscopic approach; around 100 parasites/μL. Despite these ideal features, some deviations might arise in a practical use and their evaluation is part of this thesis work. Finally, in view of an on-field use in endemic and poor developing realities, the system should be as simple as possible in order not to require any specialized personnel. The capability of the device of performing the magnetophoretic separation has been tested by means of capture experiments. The majority of these experiments have been conducted with treated bovine red blood cells (t-RBCs) stained with a fluorescent marker based on the green fluorescent protein (GFP). The treatment consists of turning the oxyhemoglobin, which has a diamagnetic property, into methemoglobin (i.e. the oxidized form of hemoglobin) that, instead, is paramagnetic [62]. The geometry of attraction required an inverted microscope and the fluorescence allowed to accurately count the cells, evaluating therefore the “capture efficiency” as a ratio between the attracted t-RBCs and the total t-RBCs in each field of view. The evaluation of this capture efficiency ratio, allowed to follow the dynamic of attraction over time and in different conditions. A comparison between the bovine and the human blood has been firstly performed, showing an identical behavior, ensuring therefore a significance of the further experiments. The experimental trials aimed at determining the dynamic of attraction in different buffer solutions (i.e. different dilutions) and compositions (i.e. with or without healthy RBCs matrix with different hematocrit levels). A dilution of 10 times in PBS (Phosphate Buffer Saline) has been chosen since it allowed a reliable count of the stained RBCs. One of the goals of these experiments was the determination of the best layout for the Ni array among different geometries. The best geometry composes of a hexagonal layout of Ni cilinders, fabricated into the silicon substrate, of 20µm height, a diameter of 40µm and a relative spacing between their centres of 160µm. This geometry has highlighted an achievable level of capture efficiency, within 15 minutes, of around 75% in a suspension of RBCs in plasma and PBS for a total hematocrit value of 4% (Hct), where the volumetric fraction of infected RBCs was 0,02% (Ft). This capture efficiency has been improved by applying an active vibrating stimulus to the system, reaching a value of 85% after only 5 minutes. In the same conditions, the false positive outcomes (a-specific capture efficiency) has been evaluated to be null in the first 10 minutes and few percentage points in the following 5 minutes, probably due to the fluorescent dye locally released in solution. A final experiment has demonstrated the possibility of working also on the “whole blood” since the platelets and the white blood cells does not affect the capture outcome that much. To conclude, these tests have proven the feasibility of the magnetophoretic separation with a specificity of 100% in the first 5 minutes. If confirmed also for real human infected blood, this aspect would represent a great advantage with respect to the state of the art of microscopic technique and RDTs. Based on the previous results, a perfect chip active area, as well as the optimal geometrical layout for the connection to the reusable module, have been designed. Several detection experiments have been therefore conducted with both treated bovine erythrocytes and hemozoin crystals. The former gave a positive result with an inverted capture geometry (i.e. sedimentation) whereas, the latter were positively detected also in the direct geometry, providing a calibration curve up to a concentration of crystals equal to 1,25µg/µL. In both cases, however, an improvement in the resolution of the electronic readout system needs to be accomplished for achieving the limit of detection of the current RDTs and microscope-based diagnostic technologies. This thesis work concerns the development of the silicon chip fabrication process as well as the experiments of magnetophoretic capture conducted in order to optimize the disposable chip, in terms of concentrators and electrodes, the geometrical layout and the connection to the reusable module. Moreover, the theoretical limit of detection (LoD) of this technology is investigated and some preliminar impedance detection measurement are performed. The silicon chip has been realized at PoliFab, the micro- and nano-fabrication facility of “Politecnico di Milano” with the standard photolithographic processes, exploiting cleanroom microfabrication equipment and techniques. The experimental activities have been performed under the supervision of Prof. Riccardo Bertacco, head of the Nanomagnetism group of the Department of Physics. The bovine and human blood samples used in the experiments, along with the required preparation, were provided by “µBS lab” of “Politecnico di Milano” and some of the magnetophoretic validation experiments where conducted in the “Mechanobiology lab”. The integrated electronic board that should allow the differential measure directly onto the module has been designed, in parallel, in the framework of another MS thesis work of the "I3N group” of “Politecnico di Milano”. The results achieved pave the way for a further experimental campaign at “Luigi Sacco” hospital in Milan in view of a second and most relevant on-field experimental campaign directly in the endemic area of Youndeè (capital of Camerun) at the “Centre Medical jean Zoa” hospital.File | Dimensione | Formato | |
---|---|---|---|
2018_12_Giuliani.pdf
non accessibile
Descrizione: Tesi Giuliani Enrico
Dimensione
38.89 MB
Formato
Adobe PDF
|
38.89 MB | Adobe PDF | Visualizza/Apri |
I documenti in POLITesi sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
https://hdl.handle.net/10589/147241