User login
Plasmodium falciparum
Credit: CDC/Mae Melvin
A semi-automated system may allow healthcare professionals to diagnose malaria infection with more than 90% accuracy.
With this system, a computer algorithm analyzes red blood cells from a digitized slide, ranks them according to the likelihood of infection, and presents the user with more than 100 images from which to make a diagnosis.
Johan Lundin, MD, PhD, of the Institute for Molecular Medicine Finland, and his colleagues described the system in PLOS ONE.
The researchers noted that high-quality microscopy is still the most accurate method for detecting malaria infection. However, as microscopy can be time-consuming, the team wanted to streamline the process by developing a semi-automated system.
“We are not suggesting that the whole malaria diagnostic process could or should be automated,” Dr Lundin said. “Rather, our aim is to develop methods that are significantly less labor-intensive than the traditional ones and have a potential to considerably increase the throughput in malaria diagnostics.”
The group’s method is based on computer vision algorithms similar to those used in facial recognition software. First, a thin layer of blood smeared on a microscope slide is digitized. Then, a computer algorithm analyzes more than 50,000 red blood cells per sample and ranks them according to the probability of malaria infection.
Next, the program creates a panel containing images of more than a hundred cells that are the most likely to be infected and presents that panel to the user. The final diagnosis is made by a healthcare professional based on the images.
To test this system, Dr Lundin and his colleagues used a set of samples from 19 patients already diagnosed with malaria and 12 control subjects. From each sample, the researchers created a digitized slide, and the system generated 128 images of the most probable parasite candidate regions.
Two expert microscopists viewed the images on a tablet computer to determine whether a subject was infected with Plasmodium falciparum.
The diagnostic sensitivity was 90% with one viewer and 95% for the other. The specificity was 100% for both viewers.
Based on these results, the researchers said this system has the potential to increase the throughput in malaria diagnostics. However, it does require some tweaking, and the team would like to expand its capabilities.
“The equipment needed for digitization of the samples is a challenge in developed countries,” said study author Nina Linder, MD, PhD, also of the Institute for Molecular Medicine Finland. “In the next phase of our project, we will test the system in combination with inexpensive mobile microscopy devices that our group has also developed.”
“There is also a strong need for fast and accurate methods for measuring the malaria parasite load in a sample,” she added. “Various malaria drug screening programs are underway, and the parasite load in a large number of samples needs to be quantified for determining the efficacy of potential drugs. We are further developing the computer algorithms used in this study to meet this need as well.”
Lastly, the researchers said this system could be applied in various other fields of medicine. In addition to other infectious diseases such as tuberculosis, the group is planning to test the system’s utility in cancer diagnosis. 
Plasmodium falciparum
Credit: CDC/Mae Melvin
A semi-automated system may allow healthcare professionals to diagnose malaria infection with more than 90% accuracy.
With this system, a computer algorithm analyzes red blood cells from a digitized slide, ranks them according to the likelihood of infection, and presents the user with more than 100 images from which to make a diagnosis.
Johan Lundin, MD, PhD, of the Institute for Molecular Medicine Finland, and his colleagues described the system in PLOS ONE.
The researchers noted that high-quality microscopy is still the most accurate method for detecting malaria infection. However, as microscopy can be time-consuming, the team wanted to streamline the process by developing a semi-automated system.
“We are not suggesting that the whole malaria diagnostic process could or should be automated,” Dr Lundin said. “Rather, our aim is to develop methods that are significantly less labor-intensive than the traditional ones and have a potential to considerably increase the throughput in malaria diagnostics.”
The group’s method is based on computer vision algorithms similar to those used in facial recognition software. First, a thin layer of blood smeared on a microscope slide is digitized. Then, a computer algorithm analyzes more than 50,000 red blood cells per sample and ranks them according to the probability of malaria infection.
Next, the program creates a panel containing images of more than a hundred cells that are the most likely to be infected and presents that panel to the user. The final diagnosis is made by a healthcare professional based on the images.
To test this system, Dr Lundin and his colleagues used a set of samples from 19 patients already diagnosed with malaria and 12 control subjects. From each sample, the researchers created a digitized slide, and the system generated 128 images of the most probable parasite candidate regions.
Two expert microscopists viewed the images on a tablet computer to determine whether a subject was infected with Plasmodium falciparum.
The diagnostic sensitivity was 90% with one viewer and 95% for the other. The specificity was 100% for both viewers.
Based on these results, the researchers said this system has the potential to increase the throughput in malaria diagnostics. However, it does require some tweaking, and the team would like to expand its capabilities.
“The equipment needed for digitization of the samples is a challenge in developed countries,” said study author Nina Linder, MD, PhD, also of the Institute for Molecular Medicine Finland. “In the next phase of our project, we will test the system in combination with inexpensive mobile microscopy devices that our group has also developed.”
“There is also a strong need for fast and accurate methods for measuring the malaria parasite load in a sample,” she added. “Various malaria drug screening programs are underway, and the parasite load in a large number of samples needs to be quantified for determining the efficacy of potential drugs. We are further developing the computer algorithms used in this study to meet this need as well.”
Lastly, the researchers said this system could be applied in various other fields of medicine. In addition to other infectious diseases such as tuberculosis, the group is planning to test the system’s utility in cancer diagnosis.
Plasmodium falciparum
Credit: CDC/Mae Melvin
A semi-automated system may allow healthcare professionals to diagnose malaria infection with more than 90% accuracy.
With this system, a computer algorithm analyzes red blood cells from a digitized slide, ranks them according to the likelihood of infection, and presents the user with more than 100 images from which to make a diagnosis.
Johan Lundin, MD, PhD, of the Institute for Molecular Medicine Finland, and his colleagues described the system in PLOS ONE.
The researchers noted that high-quality microscopy is still the most accurate method for detecting malaria infection. However, as microscopy can be time-consuming, the team wanted to streamline the process by developing a semi-automated system.
“We are not suggesting that the whole malaria diagnostic process could or should be automated,” Dr Lundin said. “Rather, our aim is to develop methods that are significantly less labor-intensive than the traditional ones and have a potential to considerably increase the throughput in malaria diagnostics.”
The group’s method is based on computer vision algorithms similar to those used in facial recognition software. First, a thin layer of blood smeared on a microscope slide is digitized. Then, a computer algorithm analyzes more than 50,000 red blood cells per sample and ranks them according to the probability of malaria infection.
Next, the program creates a panel containing images of more than a hundred cells that are the most likely to be infected and presents that panel to the user. The final diagnosis is made by a healthcare professional based on the images.
To test this system, Dr Lundin and his colleagues used a set of samples from 19 patients already diagnosed with malaria and 12 control subjects. From each sample, the researchers created a digitized slide, and the system generated 128 images of the most probable parasite candidate regions.
Two expert microscopists viewed the images on a tablet computer to determine whether a subject was infected with Plasmodium falciparum.
The diagnostic sensitivity was 90% with one viewer and 95% for the other. The specificity was 100% for both viewers.
Based on these results, the researchers said this system has the potential to increase the throughput in malaria diagnostics. However, it does require some tweaking, and the team would like to expand its capabilities.
“The equipment needed for digitization of the samples is a challenge in developed countries,” said study author Nina Linder, MD, PhD, also of the Institute for Molecular Medicine Finland. “In the next phase of our project, we will test the system in combination with inexpensive mobile microscopy devices that our group has also developed.”
“There is also a strong need for fast and accurate methods for measuring the malaria parasite load in a sample,” she added. “Various malaria drug screening programs are underway, and the parasite load in a large number of samples needs to be quantified for determining the efficacy of potential drugs. We are further developing the computer algorithms used in this study to meet this need as well.”
Lastly, the researchers said this system could be applied in various other fields of medicine. In addition to other infectious diseases such as tuberculosis, the group is planning to test the system’s utility in cancer diagnosis.